JJOD-2367; No.

of Pages 19

journal of dentistry xxx (2014) xxxxxx

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.intl.elsevierhealth.com/journals/jden

Periodontally compromised vs. periodontally

healthy patients and dental implants: A systematic
review and meta-analysis

Bruno Ramos Chrcanovic a,*, Tomas Albrektsson a,b, Ann Wennerberg a

a
Department of Prosthodontics, Faculty of Odontology, Malmo University, Malmo, Sweden
b
Department of Biomaterials, Goteborg University, Goteborg, Sweden

article info abstract

Article history: Objectives: To test the null hypothesis of no difference in the implant failure rates, postop-
Received 22 April 2014 erative infection, and marginal bone loss for the insertion of dental implants in periodon-
Received in revised form tally compromised patients (PCPs) compared to the insertion in periodontally healthy
28 August 2014 patients (PHPs), against the alternative hypothesis of a difference.
Accepted 25 September 2014 Methods: An electronic search without time or language restrictions was undertaken in
Available online xxx March 2014. Eligibility criteria included clinical human studies, either randomized or not.
Results: 2768 studies were identified in the search strategy and 22 studies were included. The
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 rate 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
Marginal bone loss significant effects of dental implants inserted in PCPs on the occurrence of postoperative
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.

Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
JJOD-2367; No. of Pages 19
2 journal of dentistry xxx (2014) xxxxxx
1. Introduction
In an attempt to decrease implant failure rates, more
attention is being placed on understanding the etiologic
and risk factors that lead to the failure of dental implants.1
The question if patients with a history of periodontitis are
more at risk for peri-implant disease has received increasing
attention in the last years.2 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.3 A history of treated periodontitis does not seem to
adversely affect implant survival rates over short times of
follow-up.4 A small number of periodontal maintenance
patients seem to be refractory to treatment and go on to
experience continued and significant tooth loss. These
subjects also have a high level of implant complications
and failure.5 However, the finding that titanium implants are
but foreign bodies have resulted in a general questioning
whether periodontitis and peri-implantitis are at all related
forms of disease.6
Some clinicians assume that periodontally compromised
patients (PCPs) present a potentially higher risk for
implant failure than healthy individuals. The reason
for this assumption is that a similar pathological
bacterial flora forms around diseased teeth and diseased
implants, though with some differences in partially and
completely edentulous patients.7 Implants are rapidly
colonized by indigenous periodontal pathogens in partially
dentate patients harbouring periodontal lesions.7 Moreover,
long-term outcomes demonstrated that implants in non-
smoking PCPs previously treated for periodontitis
were more prone to developing marginal bone loss com-
pared with those in PHPs.8 These results were obtained
despite the fact that all patients were regularly enrolled in
and were compliant with a supporting periodontal therapy
(SPT) programme over 10 years.8 Fardal and Linden5
observed that smoking, stress and a family history of
periodontal disease were identified as factors associated
with a refractory outcome, and these variables remained
significant after multivariate analysis. Another study
showed that marginal bone level at 10 years was signifi-
cantly associated with smoking, implant location, full-
mouth probing attachment levels, and change, over time,
in full-mouth probing pocket depths.9 Having said this,
a recent investigation demonstrated significantly
different mRNA signatures between periodontitis and
peri-implantitis.10
Therefore, a pertinent question in relation to implant
therapy in patients susceptible to periodontitis is whether
these patients may also show an elevated risk for peri-
implant tissue destruction. Thus, the aim of this meta-
analysis is to compare the survival rate of dental implants,
postoperative infection, and marginal bone loss of dental
implants inserted in PCPs and in periodontally healthy
patients (PHPs). The present study presents a more detailed
analysis of the influence of periodontal disease on the
implant failure rates, previously assessed in a published
systematic review.1
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
2. Materials and methods
This study followed the PRISMA Statement guidelines.11 A
review protocol does not exist.
2.1. Objective
The purpose of the present review was to test the null
hypothesis of no difference in the implant failure rates,
postoperative infection, and marginal bone loss for the
insertion of dental implants in PCPs compared to the insertion
in PHPs, against the alternative hypothesis of a difference.
2.2. Search strategies
An electronic search without time or language restrictions was
undertaken in March 2014 in the following databases: PubMed,
Web of Science, and the Cochrane Oral Health Group Trials
Register. The following terms were used in the search strategy
on PubMed:
(dental implant [Text Word]) AND periodontal disease [Text
Word]
(dental implant [Text Word]) AND periodontitis [Text Word]
The following terms were used in the search strategy on
Web of Science, in all databases, refined by selecting the term
dentistry oral surgery medicine in the filter research area:
(dental implant [Topic]) AND periodontal disease [Topic]
(dental implant [Topic]) AND periodontitis [Topic]
The following terms were used in the search strategy on the
Cochrane Oral Health Group Trials Register:
(dental implant OR dental implant failure OR dental
implant survival OR dental implant success AND (peri-
odontal disease OR periodontitis))
A manual search of dental implants-related journals,
including British Journal of Oral and Maxillofacial Surgery,
Clinical Implant Dentistry and Related Research, Clinical Oral
Implants Research, European Journal of Oral Implantology,
Implant Dentistry, International Journal of Oral and Maxillo-
facial Implants, International Journal of Oral and Maxillofacial
Surgery, International Journal of Periodontics and Restorative
Dentistry, International Journal of Prosthodontics, Journal of
Clinical Periodontology, Journal of Dental Research, Journal of
Dentistry, Journal of Oral Implantology, Journal of Craniofacial
Surgery, Journal of Cranio-Maxillofacial Surgery, and Journal
of Maxillofacial and Oral Surgery, Journal of Oral and
Maxillofacial Surgery, Journal of Oral Rehabilitation, Journal
of Periodontology, and Oral Surgery Oral Medicine Oral
Pathology Oral Radiology and Endodontology was also
performed.
The reference list of the identified studies and the
relevant reviews on the subject were also scanned for
possible additional studies. Moreover, online databases
providing information about clinical trials in progress were
JJOD-2367; No. of Pages 19
journal of dentistry xxx (2014) xxxxxx
checked (https://clinicaltrials.gov/; www.centerwatch.com/
clinicaltrials; www.clinicalconnection.com).
2.3. Inclusion and exclusion criteria
Eligibility criteria included clinical human studies, either
randomized or not, comparing implant failure rates in
any group of patients receiving dental implants that are
being inserted in PCPs compared to their insertion in PHPs.
For this review, implant failure represents the complete loss
of the implant. Exclusion criteria were case reports,
technical reports, animal studies, in vitro studies, and
review papers.
2.4. Study selection
The titles and abstracts of all reports identified through the
electronic searches were read independently by the three
authors. For studies appearing to meet the inclusion criteria,
or for which there were insufficient data in the title and
abstract to make a clear decision, the full report was obtained.
Disagreements were resolved by discussion between the
authors.
2.5. Quality assessment
Quality assessment of the studies was executed according to
the NewcastleOttawa scale (NOS).12 The NOS calculates the
study quality on the basis of 3 major components: selection,
comparability, and outcome for cohort studies. It assigns a
maximum of 4 stars for selection, a maximum of 2 stars for
comparability, and a maximum of 3 stars for outcome.
According to that quality scale, a maximum of 9 stars/points
can be given to an observational study, and this score
represents the highest quality, where six or more points were
considered high quality.
2.6. Data extraction and meta-analysis
From the studies included in the final analysis, the following
data was extracted (when available): year of publication, study
design, unicenter or multicenter study, number of patients,
patients age, follow-up, days of antibiotic prophylaxis, mouth
rinse, implant healing period, failed and placed implants,
postoperative infection (reported incidence of peri-implanti-
tis), marginal bone loss, implant surface modification,
periodontal disease definitions, periodontal therapy adopted,
use of grafting procedures, and presence of smokers among
the patients. Contact with authors for possible missing data
was performed.
Implant failure and postoperative infection were the
dichotomous outcome measures evaluated. Weighted mean
differences were used to construct forest plots of marginal
bone loss, a continuous outcome. The statistical unit for the
outcomes was the implant. Whenever outcomes of interest
were not clearly stated, the data were not used for analysis.
The I2 statistic was used to express the percentage of the
total variation across studies due to heterogeneity, with 25%
corresponding to low heterogeneity, 50% to moderate and
75% to high. The inverse variance method was used for
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
3
random-effects or fixed-effects model. Where statistically
significant (P < 0.10) heterogeneity is detected, a random-
effects model was used to assess the significance of
treatment effects. Where no statistically significant hetero-
geneity is found, analysis was performed using a fixed-
effects model.13 The estimates of relative effect for
dichotomous outcomes were expressed in risk ratio (RR)
and in mean difference (MD) in millimetres for continuous
outcomes, both with a 95% confidence interval (CI). The
degree of statistical significance was considered P < 0.05.
Only if there were studies with similar comparisons
reporting the same outcome measures was meta-analysis
to be attempted. In the case where no events (or all events)
are observed in both groups the study provides no
information about relative probability of the event and is
automatically omitted from the meta-analysis. In this
(these) case(s), the term not estimable is shown under
the column of RR of the forest plot table. The software used
here automatically checks for problematic zero counts, and
adds a fixed value of 0.5 to all cells of study results tables
where the problems occur.
A funnel plot (plot of effect size vs. standard error) will be
drawn. Asymmetry of the funnel plot may indicate publication
bias and other biases related to sample size, although the
asymmetry may also represent a true relationship between
trial size and effect size.
The data were analyzed using the statistical software
Review Manager (version 5.2.11, The Nordic Cochrane
Centre, The Cochrane Collaboration, Copenhagen, Denmark,
2014).
3. Results
3.1. Literature search
The study selection process is summarized in Fig. 1. The
search strategy resulted in 2768 papers. Two combinations of
terms were used for PubMed and Web of Science, which
resulted in a number of 360 duplicates. The three reviewers
independently screened the abstracts for those articles related
to the focus question. The initial screening of titles
and abstracts resulted in 34 full-text papers; 2374 were
excluded for not being related to the topic. The full-text
reports of the remaining 34 articles led to the exclusion of 12
because they did not meet the inclusion criteria (6 did not
inform of the number of implants per group, 2 were not
evaluating implant failures, 2 had earlier follow-up of the
same study, 1 had the same study published in another
journal, 1 compared between patients with different types of
periodontitis and not between PCPs and PHPs). Additional
hand-searching of the reference lists of selected studies did
not yield additional papers. Thus, a total of 22 publications
were included in the review.
3.2. Description of the studies
Detailed data of the 22 included studies are listed in Tables 1
and 2. Ten CCTs,2,1422 and twelve retrospective analy-
ses5,8,2332 were included in the meta-analysis.
 4

JJOD-2367; No. of Pages 19

Table 1 Detailed data of the included studies.
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

Study Published Study design Patients (n) Patients age range Follow-up visits Antibiotics/ Healing period/ Failed/placed Implant
(number per group) (average) (years) (or range) mouth rinse loading implants (n) failure
(days) rate (%)
Rosenquist and 1996 RA (unicenter) 51 (31, G1; 20, G2) 1672 (32.9) Mean 30.5 months 10/NM 6 months (maxilla) 5/62 (G1) 8.06 (G1)
Grenthe [23] (range 167) 4 months (mandible) 2/47 (G2) 4.26 (G2)
Brocard et al. [24] 2000 RA (multicenter) 440 (147, G1; 293, G2) 1690 (53) 7 years NM 6 months 39/375 (G1) 10.40 (G1)
29/647 (G2) 4.48 (G2)
Polizzi et al. [14] 2000 CCT (multicenter) 143 (NM) NM (47, females) 1, 3, and 5 years NM 6 months (maxilla) 14/98 (G1) 14.29 (G1)
NM (40, males) 34 months (mandible) 3/166 (G2) 1.81 (G2)
Watson et al. [15] 2000 CCT (unicenter) 26 (7, G1; 19, G2) 2263 (NM) 4 years NM delayed 0/7 (G1) 0 (G1)
0/26 (G2) 0 (G2)
Hardt et al. [25] 2002 RA (unicenter) 50 (25, G1; 25, G2) NM (53.5, G1) 5 years NM NM 8/100 (G1) 8 (G1)
NM (57.3, G2) 3/92 (G2) 3.26 (G2)
Karoussis et al. [16] 2003 CCT (unicenter) 53 (8, G1; 45, G2) NM 10 years NM 46 months 2/21 (G1) 9.52 (G1)
3/91 (G2) 3.30 (G2)

journal of dentistry xxx (2014) xxxxxx

Evian et al. [26] 2004 RA (unicenter) 149 (77, G1; 72, G2) NM 184030 days NM 46 months 16/77 (G1) 20.78 (G1)
6/72 (G2) 8.33 (G2)
Rosenberg et al. [27] 2004 RA (unicenter) 334 (151, G1; 183, G2) NM (61.1, G1) 13 years 710/28 days 59 months 86/923 (G1) 9.32 (G1)
NM (49.5, G2) 37/588 (G2) 6.29 (G2)
Mengel and 2005 CCT (unicenter) 39 (27, G1a; 12, G2) 1959 (3234, G1; 31, G2) 3 years NM 6 months (maxilla) 2/120 (G1a) 1.67 (G1)
Flores-de-Jacoby [17] 3 months (mandible) 0/30 (G2) 0 (G2)
Wagenberg 2006 RA (unicenter) 891 (NM) 1494 (57.9) Mean 71 months 12/NM 6 months (maxilla) 10/122 (G1) 8.20 (G1)
and Froum [28] (range 12193) 3 months (mandible) 67/1803 (G2) 3.72 (G2)
(64 immediately loaded)
Mengel et al. [18] 2007 CCT (unicenter) 17 (9, G1; 8, G2) 1959 (34, G1; 31, G2) Every 3 months NM 6 months (maxilla) 1/41 (G1) 2.44 (G1)
over a 3-year period 3 months (mandible) 0/13 (G2) 0 (G2)
Fardal and Linden [5] 2008 RA (multicenter) 16 (14, G1; 2, G2) NM (48) Mean 13.4 years NM NM 17/68 (G1) 25 (G1)
(range 819) 0/2 (G2) 0 (G2)
Gatti et al. [19] 2008 CCT (multicenter) 62 (33, G1b; 29, G2) 3585 (56, G1) 5 years NM/14 Ranging from immediate 2/155 (G1b) 1.29 (G1)
1861 (40, G2) to 11 months 0/72 (G2) 0 (G2)
De Boever et al. [20] 2009 CCT (unicenter) 194 (84, G1a; 110, G2) 2080 (53.8) Mean 46.8 (G1) and 0/7 5.9  2.1 months 16/252 (G1a) 6.35 (G1)
48.1 months (G2) 8/261 (G2) 3.07 (G2)
Anner et al. [29] 2010 RA (unicenter) 475 (311, G1; 164, G2) NM (52) Mean 30 months NM NM 61/1171 (G1) 5.21 (G1)
(range 1114) 16/455 (G2) 3.52 (G2)
Gianserra et al. [30] 2010 RA (multicenter) 1477 (1281, G1c; 196, G2) 1885 (5054, G1) 5 years 35/14 Ranging from immediate 204/5346 (G1c) 3.82 (G1)
1785 (29.9, G2) to 9 months 15/497 (G2) 3.02 (G2)
Matarasso et al. [8] 2010 RA (multicenter) 80 (40, G1; 40, G2) NM (46.548.1) 10 years NM 46 months 4/40 (G1) 10 (G1)
2/40 (G2) 5 (G2)
Simonis et al. [31] 2010 RA (unicenter) 55 (NM) 2988 (68.7) 1016 years 710/NM 34 months 5/34 (G1) 14.71 (G1)
9/97 (G2) 9.28 (G2)
Aglietta et al. [32] 2011 RA (multicenter) 40 (20, G1; 20, G2) NM (51) 10 years NM 46 months 3/20 (G1) 15 (G1)
1/20 (G2) 5 (G2)
Levin et al. [21] 2011 CCT (unicenter) 717 (434, G1d; 283, G2) NM (54, G1) Mean 54 months NM NM 70/1512 (G1d) 4.63 (G1)
NM (46, G2) (up to 144 months) 23/747 (G2) 3.08 (G2)
Roccuzzo et al. [2] 2012 CCT (unicenter) 101 (73, G1b; 28, G2) NM 10 years NM 36 months 16/185 (G1b) 8.65 (G1)
2/61 (G2) 3.28 (G2)
Roccuzzo et al. [22] 2014 CCT (unicenter) 123 (91, G1b; 32, G2) NM (53, G1) 10 years NM 612 weeks 6/198 (G1b) 3.03 (G1)
NM (43, G2) 0/54 (G2) 0 (G2)
 JJOD-2367; No. of Pages 19
Study Published Study design P value Postoperative P value (for Marginal Implant surface Periodontal Periodontal Observations
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

(for failure infection postoperative bone loss modification disease therapy

rate) infection) (mean  SD) (mm) (brand) definitions
Rosenquist and 1996 RA (unicenter) NM 4 (G1) NM NM Turned Extraction indication NM All implants in
Grenthe [23] 1 (G2) (Nobelpharma, had been fresh extraction
Nobelpharma AB, periodontitis sockets, use of
Goteborg, Sweden membranes in 5
patients
Brocard 2000 RA (multicenter) NM NM NM NM TPS (hollow screws, Prior to implant Prior to implant 132 smokers, 66
et al. [24] n = 464; solid placement, some placement, the G1 bruxers, 177
screws, n = 251; patients were treated patients were sites with GBR
hollow cylinders, for periodontal treated for
n = 307; ITI, disease periodontal disease.
Straumann, This involved a
Waldenburg, hygienic phase
Switzerland) consisting of
scaling, root

journal of dentistry xxx (2014) xxxxxx

planning, and oral
hygiene
instructions,
followed in some
cases by periodontal
surgery. All patients
were enrolled in a
periodontal
maintenance
programme with
regular professional
plaque control
Polizzi 2000 CCT (multicenter) NM NM NM NM Turned (Branemark, Periodontitis cited as NM 146 implants in
et al. [14] Nobel Biocare AB, a reason for tooth fresh extraction
Goteborg, Sweden) extraction, history sockets,
of periodontitis before membranes
tooth extraction used in 64
implants, 8
grafts
Watson 2000 CCT (unicenter) NM NM NM NM Hydroxyapatitie- Chronic periodontitis: NM Smokers were
et al. [15] coated (Calcitek pockets 4 mm and included, but the
omniloc, Carlsbad, radiographic bone exact number
USA) loss was not
informed, no
grafts, only
single-tooth
restorations

5
 6

JJOD-2367; No. of Pages 19

systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

Table 1 (Continued )
Study Published Study design P value Postoperative P value (for Marginal Implant surface Periodontal Periodontal Observations
(for failure infection postoperative bone loss modification disease therapy
rate) infection) (mean  SD) (mm) (brand) definitions
Hardt 2002 RA (unicenter) NM NM NM 2.2  0.8 (G1) Turned (Branemark, An overall descriptor NM Fixed partial
et al. [25] 1.7  0.8 (G2) Nobel Biocare AB, of the patients dentures in the
Goteborg, Sweden) experience of maxillary
periodontal posterior
destruction before the segments, no
time of implant grafts

journal of dentistry xxx (2014) xxxxxx

therapy was
generated through
the calculation of an
age-related
periodontal bone loss
score
Karoussis 2003 CCT (unicenter) NM NM NM Mesial, 1.00  1.38 (G1) TPS (hollow screws, Patients having lost The patients had 28 implants
et al. [16] 0.48  1.10 (G2) ITI, Straumann, their teeth due to been treated for placed in 12
Distal, 0.94  0.73 (G1) Waldenburg, chronic periodontitis periodontal disease smokers (10 in
0.50  1.08 (G2) Switzerland) according to a G1, 18 in G2) and
comprehensive in 41 non-
treatment strategy smokers (11 in
prior to the G1, 73 in G2)
installation of
implants
Evian 2004 RA (unicenter) NM NM NM NM ? (Paragon, Zimmer Periodontal disease Periodontal Only patients
et al. [26] Dental, Carlsbad, was diagnosed if treatment was who received a
USA) probing depths were performed prior to single implant
5 mm or greater and or in conjunction
associated with with implant
radiographic signs of placement
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
 JJOD-2367; No. of Pages 19
Rosenberg 2004 RA (unicenter) NM NM NM NM Turned (Branemark, Patients were Prior to implant
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

et al. [27] Nobel Biocare AB, classified as placement, all

Goteborg, Sweden), periodontally necessary
TPS and SLA (ITI, compromised if they periodontal,
Straumann, had a history of restorative, and
Waldenburg, periodontal disease endodontic
Switzerland), TPS that resulted in tooth treatment was
(IMZ, Biomet, Irvine, loss. Patients were completed,
USA) acid-etched classified as including extraction
(Osseotite, 3i, Palm periodontally healthy of hopeless teeth
Beach Gardens, if tooth loss was not
USA), caused by periodontal
hydroxyapatite- disease and if no loss
coated (Swede-vent, of attachment (with
Screw-vent, the exception of facial
Corevent, Paragon, or lingual recession)
Encino, USA) or probing depth
greater than 34 mm

journal of dentistry xxx (2014) xxxxxx

was present at the
time of implant
placement
Mengel and 2005 CCT (unicenter) NM NM NM 1 year, 0.83  0.71 Turned (Mk II The diagnosis of All patients No smokers
Flores-de- (G1 GA) Branemark, Nobel generalized chronic underwent
Jacoby [17] 0.68  0.54 (G1 GC) Biocare AB, and aggressive periodontal surgery
0.58  0.45 (G2) Goteborg, Sweden; periodontitis was and were entered
3 years, 0.31  0.22 n = 83), acid-etched based on the into a 3-month
(G1 GA) (Osseotite, 3i American Academy of recall system, with
0.18  0.11 (G1 GC) Implant Periodontology an oral hygiene
0.12  0.08 (G2) Innovations, Palm criteria control with
Beach Gardens, motivation and
USA; n = 67) instruction where
necessary.
Subgingival scaling
with root planing
was performed at
tooth surfaces with
probing depths
>4 mm and
bleeding on probing
Wagenberg 2006 RA (unicenter) 0.02 NM NM NM Turned (Branemark, Teeth were lost All patients were All implants
and Froum Nobel Biocare AB, because of treated for their placed in fresh
[28] Goteborg, Sweden, periodontal disease periodontal disease extraction
n = 1398), acid- prior to or in sockets, bone
etched (Osseotite, 3i conjunction with grafts were
Implant their implant utilized in all
Innovations, Palm treatment cases in which
Beach Gardens, there was a
USA; n = 527) residual space
around the
implant, 13
implants in
sinus-lifts, 323
implants in
smokers

7
 8

JJOD-2367; No. of Pages 19

Table 1 (Continued )
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

Study Published Study design P value Postoperative P value (for Marginal Implant surface Periodontal Periodontal Observations
(for failure infection postoperative bone loss modification disease therapy
rate) infection) (mean  SD) (mm) (brand) definitions
Mengel 2007 CCT (unicenter) NM NM NM 1 year, 1.02  0.89 (G1) Acid-etched The diagnosis of All generalized Edentulous
et al. [18] 0.52  0.23 (G2) (Osseotite, BIOMET/ generalized aggressive patients. G1
3 years, 0.27  0.22 (G1) 3i, Palm Beach aggressive periodontitis patients were
0.19  0.23 (G2) Gardens, USA) periodontitis was patients underwent fitted with
based on the periodontal removable
American Academy of treatment and were implant-tooth
Periodontology entered into a 3- supported
criteria month recall system superstructures,
G2 patients
received either
fixed cemented
implant-

journal of dentistry xxx (2014) xxxxxx

supported
dentures in the
maxilla or
single-tooth
implants. No
smokers.
Fardal and 2008 RA (multicenter) NM 5 (G1) NM NM NM The term refractory Patients received Even though the
Linden [5] 0 (G2) periodontal disease initial periodontal number of
has been applied to therapy, followed by smokers was
such individuals who at least 8 years of informed for the
are characterized by maintenance whole sample of
continued treatment in the the study, only
degeneration of the specialist practice the patients who
periodontium despite received
ongoing sanative, implants were
surgical and/or considered here,
pharmacological and the number
therapy of smokers
among these
patients was not
informed
Gatti et al. [19] 2008 CCT (multicenter) NM 4 (G1) NM 2.57  1.06 (G1 GA) Several (Nobel The periodontal At the first visit, the 39 implants in
0 (G2) 2.72  0.44 (G1 GC) Biocare, conditions were periodontal grafts, 14
1.24  1.09 (G2) Gothenburg, assessed using a conditions were smokers (8 in G1,
Sweden; Zimmer modification of the assessed using a 6 in G2)
Dental, Carlsbad, Periodontal Screening modification of the
USA; Mathys, and Recording index Periodontal
Bettlach, Screening and
Switzerland; Recording index,
Straumann, and subsequently
Waldenburg, periodontal therapy
Switzerland;
Dentsply Friadent,
Mannheim,
Germany)
 JJOD-2367; No. of Pages 19
De Boever 2009 CCT (unicenter) NM NM NM Mesial, 0.28  0.7 TPS and SLA (ITI, Periodontally Before implant 11.4% of the
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

et al. [20] (G1 + G2) Straumann, susceptible patients placement, all patients were
Distal, 0.24  0.6 Waldenburg, with tooth loss due to patients received, if smokers (13
(G1 + G2) Switzerland; n = 259, periodontal disease necessary, smokers in G1, 9
TPS; n = 254, SLA) and patients with periodontal non- in G2), 10.3%
periodontal disease surgical and/or former smokers,
surgical therapy. 134 ridge
Periodontally augmentations
susceptible patients (75, G1; 59, G2)
were enrolled in a
strict maintenance
programme
Anner et al. 2010 RA (unicenter) 0.1498 NM NM NM NM NM 246 patients (51.7%) 49 diabetics, 63
[29] participated of a smokers
structured
supportive
periodontal
programme

journal of dentistry xxx (2014) xxxxxx

Gianserra 2010 RA (multicenter) NM NM NM NM Several (3i Biomet, Periodontal Periodontal therapy 549 smokers (486
et al. [30] Astra Tech, Camlog, conditions were (non-surgical and in G1, 63 in G2)
Friadent-Dentsply, assessed using a surgical) was
Nobel Biocare, modification of the administered as
Straumann, Sweden Periodontal Screening required
& Martina, Zimmer and Recording (PSR)
Dental) index
Matarasso 2010 RA (multicenter) NM NM NM Turned, 2.78  0.48 (G1) Turned (Branemark, The classification of The patients No smokers,
et al. [8] 1.95  0.42 (G2) Nobel Biocare AB, the patients in the received only dental
TPS, 2.32  0.41 (G1) Goteborg, Sweden; two groups was individualized implants in a
1.43  0.38 (G2) n = 40), TPS (ITI, carried out on the periodontal single-unit gap
Straumann, basis of the diagnosis treatment before
Waldenburg, reported in the implant surgery. On
Switzerland; n = 40) patients chart. the basis of the
results achieved
after the
periodontal
treatment, the
patients were
placed on an
individually tailored
maintenance care
programme
Simonis 2010 RA (unicenter) 0.327 13 (G1) 0.006 Mesial, 2.2  3.4 (G1 + G2) TPS (ITI, NM All patients were 9 smokers, only
et al. [31] 10 (G2) Distal, 2.3  3.4 (G1 + G2) Straumann, instructed on how implant-
Waldenburg, to maintain supported fixed
Switzerland; solid appropriate oral restorations
screw, n = 116; hygiene around the
hollow screw, implants and
n = 15) remaining teeth.

9
 10

JJOD-2367; No. of Pages 19

systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

Table 1 (Continued )
Study Published Study design P value Postoperative P value (for Marginal Implant surface Periodontal Periodontal Observations
(for failure infection postoperative bone loss modification disease therapy
rate) infection) (mean  SD) (mm) (brand) definitions
Aglietta 2011 RA (multicenter) NM NM NM Turned, 3.47  1.09 (G1) Turned (Branemark, Patients treated for Patients were All patients were
et al. [32] 2.65  0.31 (G2) Nobel Biocare AB, generalized chronic treated for smokers, only
TPS, 3.77  1.43 (G1) Goteborg, Sweden; periodontitis periodontitis dental implants
2.51  0.31 (G2) n = 20), TPS (ITI, in a single-unit
Straumann, gap
Waldenburg,

journal of dentistry xxx (2014) xxxxxx

Switzerland; n = 20)
Levin 2011 CCT (unicenter) NM NM NM NM NM Patients were divided All periodontally 81 diabetics, 103
et al. [21] into different involved patients smokers (71 in
periodontal groups had undergone G1, 32 in G2)
according to their cause-related as
periodontal diagnosis well as corrective-
that was based on a phase periodontal
classification of interventions (if
periodontal diseases indicated) before
dental implant
placement
Roccuzzo 2012 CCT (unicenter) NM NM NM 0.98  1.22 (G1 severe) TPS (ITI, PCP received a score All patients received 18 smokers (15
et al. [2] 1.14  1.11 (G1 moderate) Straumann, (S) on the basis of the appropriate initial in G1, 3 in G2)
0.75  0.88 (G2) Waldenburg, number and depth of therapy, consisting,
Switzerland) periodontal pockets depending on the
according to the cases, in motivation,
following formula: oral hygiene
S = Number of instruction and
pockets (57 mm) + 2 scaling and root
Number of pockets planing, with the aim
(8 mm) to reduce to a
minimal level
periodontal
pathogens. Hopeless
teeth were extracted.
Periodontal surgery
was performed as
needed. Guided
tissue regeneration
was pursued, when
feasible
 JJOD-2367; No. of Pages 19
Roccuzzo 2014 CCT (unicenter) NM NM NM NM Sandblasted and PCP received a score All patients received 21 smokers (16
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

et al. [22] acid-etched (SLA, (S) on the basis of the appropriate initial in G1, 5 in G2)
Straumann, number and depth of therapy, consisting,
Waldenburg, periodontal pockets depending on the
Switzerland) according to the cases, in
following formula: motivation, oral
S = Number of hygiene instruction
pockets (57 mm) + 2 and scaling and root
Number of pockets planing, with the
(8 mm) aim to reduce to a
minimal level
periodontal
pathogens.
Hopeless teeth were
extracted.
Periodontal surgery
was performed as
needed. Guided

journal of dentistry xxx (2014) xxxxxx

tissue regeneration
was pursued, when
feasible

NM not mentioned; CCT controlled clinical trial; RCT randomized controlled trial; RA retrospective analysis; G1 group periodontitis; G2 group non- periodontitis; NP not performed; TPS titanium plasma-sprayed; GBR
guided bone regeneration; GA aggressive periodontitis group; GC chronic periodontitis group.
a
Here the implants and the patients with chronic adult periodontitis and generalized aggressive periodontitis were put together in G1.
b
Here the implants and the patients with severe chronic periodontitis and moderate chronic periodontitis were put together in G1.
c
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.

11
JJOD-2367; No. of Pages 19

12 journal of dentistry xxx (2014) xxxxxx

Fig. 1 Study screening process.

Three studies17,18,23 had a follow-up up to 3 years, one15 of 4
years, whereas 18 studies2,5,8,14,16,1922,2432 had a maximum
follow-up of at least 5 years. From the studies with available
data of patients age, four23,24,28,30 included non-adults
patients. Three studies2,16,26 did not inform of the patients
age. Only four studies5,19,23,31 provided information about
postoperative infection, with 37 occurrences in a total of 184
patients receiving 537 implants. Some patients in thirteen
studies2,5,15,16,1922,24,2831 were smokers, whereas in one
study32 all patients were smokers, and three studies8,17,18
excluded smokers. One study14 inserted part of the implants in
fresh extraction sockets, whereas in the other two23,28 all
implants were inserted in fresh extraction sockets. Three
studies8,26,32 included only patients who received a single
implant. Patients were submitted to grafting procedures at the
implant site in 5 studies.14,19,20,24,28 One study18 included only
edentulous patients, and one25 inserted implants in the
maxillary posterior segments only. Any kind of periodontal
treatment previous to the implants insertion or a SPT was not
mentioned to be performed in four studies.14,15,23,25
Seven studies2,17,1922,30 included a comparison between
periodontitis of different severities. From the 22 studies
comparing PHPs and PCPs, a total of 10,927 dental implants
were inserted in PCPs, with 587 failures (5.37%), and 5881
implants were inserted in PHPs, with 226 failures (3.84%).
There were no implant failures in one study.15 From the 7
studies2,17,1922,30 comparing periodontitis of different seve-
rities, a total of 4460 dental implants were inserted in
patients with a more aggressive type of periodontitis, with
210 failures (4.71%), and 3308 implants were inserted in
patients with a less aggressive type of periodontitis, with 106
failures (3.20%). The inherent problem in a meta-analysis
such as in the present paper is that although some authors
see a difference between periodontitis and aggressive or
chronic periodontitis, others have not split the material in
this manner. Therefore, comparisons are difficult and we
will treat periodontitis as one entity in the remaining part of
the paper.
The most commonly used implants were the titanium
plasma-sprayed from Straumann (Straumann, Waldenburg,
Switzerland), in nine studies,2,8,16,19,20,24,27,31,32 and the Brane-
mark (Nobel Biocare AB, Goteborg, Sweden), in eight stud-
ies.8,14,17,19,25,27,28,32 The latter was not exclusively used in six
studies.8,17,19,27,28,32 Three studies5,21,29 did not inform what
kind of implants were used. Three studies28,29,31 informed
whether there was a statistically significant difference or not
between the implant failure rates between the PCPs and PHPs,
and only one28 found a statistical significance favouring PHPs.
Six studies20,23,27,28,30,31 provided information about the use of
prophylactic antibiotics. In one of them,20 it was informed that
antibiotics were not prescribed to any patient. Four stud-
ies19,20,27,30 provided information about the use of chlorhexi-
dine mouth rinse by the patients.
3.3. Quality assessment
All studies except one23 were high quality. The scores are
summarized in Table 2.

Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
 JJOD-2367; No. of Pages 19
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A

Table 2 Quality assessment of the studies by the Newcastle-Ottawa scale.

Study Published Selection Comparability Outcome Total
(9/9)
Representativeness Selection of Ascertainment Outcome of Comparability of Assessment Follow-up Adequacy of
of the exposed external of exposure interest not cohorts of outcome long follow-up
cohort control present at enougha
start
Main Additional
factor factor

journal of dentistry xxx (2014) xxxxxx

Rosenquist and 1996 0 $ $ $ $ 0 $ 0 0 5/9
Grenthe [23]
Brocard et al. [24] 2000 $ $ $ $ $ $ $ $ 0 8/9
Polizzi et al. [14] 2000 0 $ $ $ $ $ $ $ 0 7/9
Watson et al. [15] 2000 $ $ $ $ $ 0 $ 0 $ 7/9
Hardt et al. [25] 2002 0 $ $ $ $ 0 $ $ $ 7/9
Karoussis et al. [16] 2003 $ $ $ $ $ 0 $ $ 0 7/9
Evian et al. [26] 2004 0 $ $ $ $ $ $ $ 0 7/9
Rosenberg et al. [27] 2004 $ $ $ $ $ $ $ $ 0 8/9
Mengel and 2005 $ $ $ $ $ 0 $ 0 0 6/9
Flores-de-Jacoby [17]
Wagenberg and 2006 0 $ $ $ $ $ $ $ 0 7/9
Froum [28]
Mengel et al. [18] 2007 0 $ $ $ $ 0 $ 0 $ 6/9
Fardal and Linden [5] 2008 $ $ $ $ $ 0 $ $ 0 7/9
Gatti et al. [19] 2008 $ $ $ $ $ 0 $ $ $ 8/9
De Boever et al. [20] 2009 0 $ $ $ $ $ $ 0 0 6/9
Anner et al. [29] 2010 $ $ $ $ $ $ $ 0 0 7/9
Gianserra et al. [30] 2010 $ $ $ $ $ 0 $ $ $ 8/9
Matarasso et al. [8] 2010 $ $ $ $ $ $ $ $ $ 9/9
Simonis et al. [31] 2010 $ $ $ $ $ 0 $ $ 0 7/9
Aglietta et al. [32] 2011 $ $ $ $ $ $ $ $ $ 9/9
Levin et al. [21] 2011 $ $ $ $ $ $ $ 0 0 7/9
Roccuzzo et al. [2] 2012 0 $ $ $ $ 0 $ $ $ 7/9
Roccuzzo et al. [22] 2014 0 $ $ $ $ 0 $ $ 0 6/9
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.

13
JJOD-2367; No. of Pages 19

14 journal of dentistry xxx (2014) xxxxxx

Fig. 2 Forest plot for the event implant failure in the comparison between periodontally compromised vs. periodontally
healthy patients.

3.4. Meta-analysis analysis was also performed with a random-effects model.

In this study, a fixed-effects model was used to evaluate the
implant failure in the comparison between PCPs vs. PHPs,
since statistically significant heterogeneity was not found
(P = 0.87; I2 = 0%). The insertion of dental implants in PCPs or
PHPs statistically affected the implant failure rates
(P < 0.00001; Fig. 2), in favour of PHPs. A RR of 1.78 (95% CI
1.502.11) implies that failures when implants are inserted in
PCPs are 1.78 times likely to happen than failures when
implants are inserted in PHPs.
Since the effect size could differ depending on the research
methodology of the studies, a sensitivity analysis was
performed. When only the CCTs were pooled, a RR of 1.97
resulted (95% CI 1.382.80; heterogeneity: P = 0.53; I2 = 0%;
Fig. 3), also statistically affecting the implant failure rates
(P = 0.0002). Given the variability of the included studies
(varying lengths of follow-up, patient ages, number of
implants, classification of severity of periodontitis etc.), the
When all studies were evaluated or when only the CCTs were
pooled, the significance of the treatment effect was the same
as when the fixed-effects model was used, with the exact same
values for the RR and the 95% CI.
Only four studies5,19,23,31 provided information about
postoperative infection. A fixed-effects model was used, due
to lack of statistically significant heterogeneity (P = 0.54;
I2 = 0%). The insertion of dental implants in PCPs or PHPs
statistically affected the incidence of postoperative infections
(P = 0.0004; Fig. 4), in favour of PHPs. A RR of 3.24 (95% CI 1.69
6.21) was observed. When the analysis was performed with a
random-effects model, the values for RR and 95% CI remained
the same, as well as the significance of the treatment effect.
Five studies8,16,18,25,32 provided information about the
marginal bone loss with standard deviation, necessary for
the calculation of comparisons in continuous outcomes,
comparing PCPs and PHPs (212 implants in PCPs and 269
implants in PHPs; Fig. 5). A random-effects model was used to

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.

Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
JJOD-2367; No. of Pages 19

journal of dentistry xxx (2014) xxxxxx 15

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.

evaluate the marginal bone loss, since statistically significant
heterogeneity was found (P < 0.00001; I2 = 88%). There was
statistically significant difference (MD 0.60, 95% CI 0.330.87;
P < 0.0001) between the groups concerning the marginal bone
loss, favouring PHPs.
high value of RR together with a wide CI range (7.90, CI 95%,
2.3326.82), showing heterogeneity in comparison with the
other studies. When only the CCTs were pooled (Fig. 7), a
possible presence of publication bias is still indicated.

3.5. Publication bias 4. Discussion

The funnel plot showed asymmetry when the studies
reporting the outcome implant failure in the comparison
between PCPs vs. PHPs are analyzed, indicating possible
presence of publication bias. Seven studies2,8,14,16,23,25,32
collaborated with the asymmetry (Fig. 6), and showed a wide
CI range for RR. The study of Polizzi et al.14 was the only one
outside the triangular 95% confidence region, showing a very
Fig. 6 Funnel plot for the studies reporting the outcome
event implant failure.
Narrowing the inclusion criteria of studies increases homoge-
neity 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 random-
effects 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

Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
JJOD-2367; No. of Pages 19
16 journal of dentistry xxx (2014) xxxxxx
Fig. 7 Funnel plot for the studies reporting the outcome
event implant failure, when only the CCTs were pooled.
must be taken in the interpretation of the chi-squared test,
since it has low power in the (common) situation of a meta-
analysis when studies have small sample size or are few in
number. This means that while a statistically significant result
may indicate a problem with heterogeneity, a non-significant
result must not be taken as evidence of no heterogeneity.34
Some argue that, since clinical and methodological diversity
always occur in a meta-analysis, statistical heterogeneity is
inevitable.35 Thus, the test for heterogeneity is irrelevant to
the choice of analysis; heterogeneity will always exist whether
or not we happen to be able to detect it using a statistical test.35
The present meta-analysis showed that, regardless of how
the studies were pooled, either when the retrospective studies
and the CCTs were considered together, or when only the
CCTs were considered, there was a statistically significant
difference between PCPs and PHPs for the outcome implant
failure, in favour of PHPs. In patients in whom teeth were lost
for periodontal reasons, the disease may have decreased the
available bone following tooth extraction or resulted in the
necessity to place the implant with a more exposed surface to
achieve ideal prosthetic position. Both of these situations may
have resulted in a greater implant failure rate.28
Here it is important to mention the possible high influence
of smoking habits on the implants failure rates. Smokers were
included in 14 studies here reviewed, and the smoking habit is
considered a confounding factor in the present meta-analysis.
Tobacco smoking is considered the principal environmental
risk factor affecting the pathogenesis of periodontitis but it is
also responsible for a great number of other types of disease.
Karoussis et al.16 showed in their study that there was a
tendency for a poorer survival rate of implants in smokers vs.
nonsmokers in patients with a history of periodontitis,
indicating that the smoking patient susceptible to periodonti-
tis yields a documented higher risk for implant loss than the
non-smoking periodontitis patient or the patient not suscep-
tible to periodontitis at all. Aglietta et al.32 showed that tobacco
smokers with a history of treated periodontitis displayed
lower implant survival rates and higher marginal bone loss
rates compared with smokers PHPs, independent of other
factors such as implant type, healing modality and operator,
and despite the fact that smokers PCPs were treated for their
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
periodontal conditions before implant placement and were
regularly enrolled in a SPT.
When marginal bone loss was analyzed, there was a
statistically significant difference between PCPs and PHP,
favouring PHPs. Hardt et al.25 observed that the amount of
longitudinal peri-implant bone loss is related to pretreatment
experience of loss of periodontal bone support. The exact
relationship between periodontitis and peri-implantitis
remains unknown; we really do not know whether these
two types of disease are similar in origin at all since a
demonstrated positive correlation between PCP and implant
failure in itself does not prove such a connection. In some
studies PCPs demonstrated higher implant failure rates but
this difference did not reach statistical significance. Obviously,
treatment of periodontal infections before implant placement
would seem important to avoid placing an implant in an
infected bed. It is important to stress here that heterogeneity
was identified among the group of studies reporting marginal
bone loss. Thus, conclusions should be interpreted with
caution and the analysis can at best lead to the generation of
hypotheses.
The fact that some of these studies reviewed here have a
short follow-up is a confounding factor, even though it is hard
to define what would be considered a short follow-up period to
evaluate implant failures in PCPs. A longer follow-up period
can lead to an increase in the failure rate, especially if it is
extended beyond functional loading, because other prosthetic
factors can influence implant failure from that point onward.
This might have led to an underestimation of actual failures in
some studies. Analysis of the data disclosed different patterns
regarding the distribution of the implant losses over time in
the two categories of patients. Rosenberg et al.27 hypothesized
that one possible explanation for the difference between the
two groups in this pattern of failure is the influence of the host,
which plays an important role in the variable inflammatory
process and may be significant in patients with a history of
periodontal disease. Another possible explanation could be
related to local factors. A reduced quantity of hard tissue in the
PCP group may be related to periodontal loss prior to tooth
extraction.27
Another confounding factor is the fact that the studies used
different definitions for the presence of periodontal disease,
depending on the threshold chosen for the definition of
periodontitis, or which conjunct of characteristics may be
considered a periodontal disease, i.e., the diagnostic criteria
are less clear. Thus, a clear classification system needs to be
implemented with clinical evaluation related to a more
specific pathology. Moreover, the outcome measures were
not related to the type of periodontitis in every study. When it
was reported, there was a statistically significant difference
concerning the implant failure rates, favouring the less
aggressive type of periodontitis in comparison with the more
aggressive.
Numerous implants in some studies26 were placed at the
same time as periodontal surgical procedures were being
carried out. The influence of this co-therapy on implant
contamination during the procedure has not been investigat-
ed. At the immediate and early implant placement it can be
speculated that periodontitis-affected tissues might have had
a negative local influence due to the presence of infrabony
JJOD-2367; No. of Pages 19
journal of dentistry xxx (2014) xxxxxx
defects; this could increase the gap between bone and
implant36 or jeopardize achievement of primary stability.37
It is unknown whether textured implant surfaces may be
more vulnerable to infection than machined implant surfaces
in patients with past or present periodontal disease.26 Some
studies presented higher failure rates in PCPs when using TPS
implants. The moderate micro-roughness of most modern
implants did not seem coupled to more than 12% of peri-
implantitis when followed up for 10 years or more as indicated
in a recent review of ten different long term clinical reports of
Tioblast, SLA and TiUnite implants.38 Titanium with different
surface modifications shows a wide range of chemical,
physical properties, and surface topographies or morpholo-
gies, depending on how they are prepared and handled,3941
and it is not clear whether, in general, one surface modifica-
tion is better than another.42
Another possible limitation of some studies is that
implants were not tested for stability during some of the late
follow-up visits since many of the prostheses could not be
removed because they were permanently cemented. This
might have led to an underestimation of actual failures.43
Differences in prosthetic suprastructures including complete-
ly or partially edentulous patients in the same study are
variables that must also be taken into account. The small
number of patients in some studies5,15,17,18,32 also counts as a
limitation. Moreover, groups were not completely comparable
at baseline in some cases.5,16,17,29,30 The potentially most
relevant differences were in terms of age and number of
implants/prosthesis. The differences in age and number of
implants could have plausible explanations, such as different
patterns of tooth loss among groups.19 Patients with a
previous history of periodontitis are likely to have lost more
teeth because of periodontal disease, and therefore require
more implants, whereas the healthy group usually includes
patients who had lost teeth through trauma or were affected
by tooth agenesia, and therefore they were likely to be younger
and require fewer implants.19
The results of the present study have to be interpreted with
caution because of its limitations. First of all, all confounding
factors may have affected the long-term outcomes and not
just the presence or not of periodontal disease, and the impact
of these variables on the implant survival rate, postoperative
infection and marginal bone loss is difficult to estimate if these
factors are not identified separately between the two different
procedures in order to perform a meta-regression analysis.
The lack of control of the confounding factors limited the
potential to draw robust conclusions. Second, due to the
retrospective design of some studies the classification of the
patients with respect to their experience of periodontal
disease could be based only on preoperative radiographic
data describing the amount of bone support at remaining
teeth, since clinical data regarding the periodontal conditions
at time for implant therapy or at subsequent follow-ups are
not retrievable.4446 Third, there are no RCTs in the analysis,
and potential biases are likely to be greater for non-random-
ized studies compared with RCTs.
The authors suggest that more research is needed on the
history of bone tissue loss prior to implant placement in
patients classified as periodontally compromised to evalu-
ate the local factors affecting implant failure in these patients.
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
17
Due to the multifaceted aspects of any infectious disease such
as periodontitis, any correlations between this disease and
peri-implantitis need not necessarily indicate that bone loss
around teeth and implants is dependent on the same type of
disease.
5. Conclusion
The results of the present systematic review should be
interpreted with caution due to the presence of uncontrolled
confounding factors in the included studies, none of them
randomized. Within the limitations of the existing investiga-
tions, the present study suggests that an increased suscepti-
bility for periodontitis may also translate to an increased
susceptibility for implant loss, loss of supporting bone, and
postoperative infection.
Acknowledgements
This work was supported by CNPq, Conselho Nacional de
Desenvolvimento Cientfico e Tecnologico Brazil. We would
like to thank Dr. Ricardo Trindade.
references
1. Chrcanovic BR, Albrektsson T, Wennerberg A. Reasons for
failures of oral implants. Journal of Oral Rehabilitation
2014;41:44376.
2. Roccuzzo M, Bonino F, Aglietta M, Dalmasso P. Ten-year
results of a three arms prospective cohort study on implants
in periodontally compromised patients. Part 2: clinical
results. Clinical Oral Implants Research 2012;23:38995.
3. Ong CTT, Ivanovski S, Needleman IG, Retzepi M, Moles DR,
Tonetti MS, et al. Systematic review of implant outcomes in
treated periodontitis subjects. Journal of Clinical Periodontology
2008;35:43862.
4. Klokkevold PR, Han TJ. How do smoking, diabetes, and
periodontitis affect outcomes of implant treatment?
International Journal of Oral and Maxillofacial Implants
2007;22(Suppl.):173202.
5. Fardal O, Linden GJ. Tooth loss and implant outcomes in
patients refractory to treatment in a periodontal practice.
Journal of Clinical Periodontology 2008;35:7338.
6. Albrektsson T, Dahlin C, Jemt T, Sennerby L, Turri A,
Wennerberg A. Is marginal bone loss around oral implants
the result of a provoked foreign body reaction? Clinical
Implant Dentistry and Related Research 2014;16:15565.
7. Apse P, Ellen RP, Overall CM, Zarb GA. Microbiota and
crevicular fluid collagenase activity in the osseointegrated
dental implant sulcus: a comparison of sites in edentulous
and partially edentulous patients. Journal of Periodontal
Research 1989;24:96105.
8. Matarasso S, Rasperini G, Siciliano VI, Salvi GE, Lang NP,
Aglietta M. A 10-year retrospective analysis of radiographic
bone-level changes of implants supporting single-unit crowns
in periodontally compromised vs. periodontally healthy
patients. Clinical Oral Implants Research 2010;21:898903.
9. Karoussis IK, Muller S, Salvi GE, Heitz-Mayfield LJA, Bragger
U, Lang NP. Association between periodontal and peri-
implant conditions: a 10-year prospective study. Clinical Oral
Implants Research 2004;15:17.
JJOD-2367; No. of Pages 19
18 journal of dentistry xxx (2014) xxxxxx
10. Becker ST, Beck-Broichsitter BE, Graetz C, Dorfer CE,
Wiltfang J, Hasler R. Peri-implantitis versus periodontitis:
functional differences indicated by transcriptome profiling.
Clinical Implant Dentistry and Related Research 2014;16:40111.
11. Moher D, Liberati A, Tetzlaff J, Altman DG, Grp P. Preferred
reporting items for systematic reviews and meta-analyses:
the PRISMA statement. Annals of Internal Medicine
2009;151:2649. W64.
12. Wells GA, Shea B, OConnell D, Peterson J, Welch V, Losos M,
et al. The NewcastleOttawa Scale (NOS) for assessing the
quality of nonrandomised studies in meta-analyses. 2000.
Available from: http://www.ohri.ca/programs/
clinical_epidemiology/oxford.asp (accessed on 27.09.14).
13. Egger M, Smith GD. Principles of and procedures for
systematic reviews. In: Egger M, Smith GD, Altman DG,
editors. Systematic reviews in health care: meta-analysis in
context. London: BMJ Books; 2003. p. 2342.
14. Polizzi G, Grunder U, Goene R, Hatano N, Henry P, Jackson
WJ, et al. Immediate and delayed implant placement into
extraction sockets: a 5-year report. Clinical Implant Dentistry
and Related Research 2000;2:939.
15. Watson CJ, Tinsley D, Sharma S. Implant complications and
failures: the single-tooth restoration. Dental Update
2000;27:358. 40, 42.
16. Karoussis IK, Salvi GE, Heitz-Mayfield LJ, Bragger U,
Hammerle CH, Lang NP. Long-term implant prognosis in
patients with and without a history of chronic periodontitis:
a 10-year prospective cohort study of the ITI Dental Implant
System. Clinical Oral Implants Research 2003;14:32939.
17. Mengel R, Flores-de-Jacoby LF. Implants in patients treated
for generalized aggressive and chronic periodontitis: a 3-
year prospective longitudinal study. Journal of Periodontology
2005;76:53443.
18. Mengel R, Kreuzer G, Lehmann KM, Flores-de-Jacoby L. A
telescopic crown concept for the restoration of partially
edentulous patients with aggressive generalized
periodontitis: a 3-year prospective longitudinal study.
International Journal of Periodontics and Restorative Dentistry
2007;27:2319.
19. Gatti C, Gatti F, Chiapasco M, Esposito M. Outcome of dental
implants in partially edentulous patients with and without
a history of periodontitis: a 5-year interim analysis of a
cohort study. European Journal of Oral Implantology
2008;1:4551.
20. De Boever AL, Quirynen M, Coucke W, Theuniers G, De
Boever JA. Clinical and radiographic study of implant
treatment outcome in periodontally susceptible and non-
susceptible patients: a prospective long-term study. Clinical
Oral Implants Research 2009;20:134150.
21. Levin L, Ofec R, Grossmann Y, Anner R. Periodontal disease
as a risk for dental implant failure over time: a long-term
historical cohort study. Journal of Clinical Periodontology
2011;38:7327.
22. Roccuzzo M, Bonino L, Dalmasso P, Aglietta M. Long-term
results of a three arms prospective cohort study on implants
in periodontally compromised patients: 10-year data around
sandblasted and acid-etched (SLA) surface. Clinical Oral
Implants Research 2014;25:110512.
23. Rosenquist B, Grenthe B. Immediate placement of implants
into extraction sockets: implant survival. International Journal
of Oral and Maxillofacial Implants 1996;11:2059.
24. Brocard D, Barthet P, Baysse E, Duffort JF, Eller P, Justumus P,
et al. A multicenter report on 1,022 consecutively placed ITI
implants: a 7-year longitudinal study. International Journal of
Oral and Maxillofacial Implants 2000;15:691700.
25. Hardt CRE, Grondahl K, Lekholm U, Wennstrom JL. Outcome
of implant therapy in relation to experienced loss of
periodontal bone support a retrospective 5-year study.
Clinical Oral Implants Research 2002;13:48894.
Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013
26. Evian CI, Emling R, Rosenberg ES, Waasdorp JA, Halpern W,
Shah S, et al. Retrospective analysis of implant survival and
the influence of periodontal disease and immediate
placement on long-term results. International Journal of Oral
and Maxillofacial Implants 2004;19:3938.
27. Rosenberg ES, Cho SC, Elian N, Jalbout ZN, Froum S, Evian
CI. A comparison of characteristics of implant failure and
survival in periodontally compromised and periodontally
healthy patients: a clinical report. International Journal of Oral
and Maxillofacial Implants 2004;19:8739.
28. Wagenberg B, Froum SJ. A retrospective study of 1925
consecutively placed immediate implants from 1988 to
2004. International Journal of Oral and Maxillofacial Implants
2006;21:7180.
29. Anner R, Grossmann Y, Anner Y, Levin L. Smoking, diabetes
mellitus, periodontitis, and supportive periodontal
treatment as factors associated with dental implant
survival: a long-term retrospective evaluation of patients
followed for up to 10 years. Implant Dentistry 2010;19:5764.
30. Gianserra R, Cavalcanti R, Oreglia F, Manfredonia MF,
Esposito M. Outcome of dental implants in patients with
and without a history of periodontitis: a 5-year pragmatic
multicentre retrospective cohort study of 1727 patients.
European Journal of Oral Implantology 2010;3:30714.
31. Simonis P, Dufour T, Tenenbaum H. Long-term implant
survival and success: a 1016-year follow-up of non-
submerged dental implants. Clinical Oral Implants Research
2010;21:7727.
32. Aglietta M, Siciliano VI, Rasperini G, Cafiero C, Lang NP, Salvi
GE. A 10-year retrospective analysis of marginal bone-level
changes around implants in periodontally healthy and
periodontally compromised tobacco smokers. Clinical Oral
Implants Research 2011;22:4753.
33. Shrier I, Boivin JF, Steele RJ, Platt RW, Furlan A, Kakuma R,
et al. Should meta-analyses of interventions include
observational studies in addition to randomized controlled
trials? A critical examination of underlying principles.
American Journal of Epidemiology 2007;166:12039.
34. Higgins JPT, Green S, Cochrane Collaboration.Cochrane
handbook for systematic reviews of interventions.
Chichester, England/Hoboken, NJ: Wiley-Blackwell; 2008.
35. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring
inconsistency in meta-analyses. British Medical Journal
2003;327:55760.
36. Carlsson L, Rostlund T, Albrektsson B, Albrektsson T.
Implant fixation improved by close fit. Cylindrical implant-
bone interface studied in rabbits. Acta Orthopaedica
Scandinavica 1988;59:2725.
37. Ivanoff CJ, Sennerby L, Lekholm U. Influence of initial
implant mobility on the integration of titanium implants
an experimental study in rabbits. Clinical Oral Implants
Research 1996;7:1207.
38. Albrektsson T, Buser D, Sennerby L. Crestal bone loss and
oral implants. Clinical Implant Dentistry and Related Research
2012;14:78391.
39. Chrcanovic BR, Pedrosa AR, Martins MD. Chemical and
topographic analysis of treated surfaces of five different
commercial dental titanium implants. Materials Research
2012;15:37282.
40. Chrcanovic BR, Leao NLC, Martins MD. Influence of
different acid etchings on the superficial characteristics
of Ti sandblasted with Al2O3. Materials Research
2013;16:100614.
41. Chrcanovic BR, Martins MD. Study of the influence of acid
etching treatments on the superficial characteristics of Ti.
Materials Research 2014;17:37380.
42. Wennerberg A, Albrektsson T. Effects of titanium surface
topography on bone integration: a systematic review.
Clinical Oral Implants Research 2009;20:17284.
JJOD-2367; No. of Pages 19

journal of dentistry xxx (2014) xxxxxx 19

43. Grondahl K, Lekholm U. The predictive value of
radiographic diagnosis of implant instability. International
Journal of Oral and Maxillofacial Implants 1997;12:5964.
44. Chrcanovic BR, Abreu MH, Freire-Maia B, Souza LN. Facial
fractures in children and adolescents: a retrospective study
of 3 years in a hospital in Belo Horizonte, Brazil. Dental
Traumatology 2010;26:26270.
45. Chrcanovic BR, Souza LN, Freire-Maia B, Abreu MH. Facial
fractures in the elderly: a retrospective study in a hospital in
Belo Horizonte, Brazil. Journal of Trauma 2010;69:E738.
46. Chrcanovic BR, Abreu MH, Freire-Maia B, Souza LN. 1,454
mandibular fractures: a 3-year study in a hospital in Belo
Horizonte, Brazil. Journal of Cranio-Maxillofacial Surgery
2012;40:11623.

Please cite this article in press as: Chrcanovic BR, et al. Periodontally compromised vs. periodontally healthy patients and dental implants: A
systematic review and meta-analysis. Journal of Dentistry (2014), http://dx.doi.org/10.1016/j.jdent.2014.09.013