Dan Zhao† Immediate dental implant placement

Yaqin Wu†
Chun Xu
into infected vs. non-infected sockets:
Fuqiang Zhang a meta-analysis

Authors’ affiliations:
Dan Zhao, Yaqin Wu, Chun Xu, Fuqiang Zhang,
Department of Prosthodontics, Shanghai Ninth
People’s Hospital, Shanghai Jiao Tong University
School of Medicine, Shanghai, China
Dan Zhao, Yaqin Wu, Chun Xu, Fuqiang Zhang,
Shanghai Key Laboratory of Stomatology, Shanghai
Research Institute of Stomatology, Shanghai, China
Corresponding authors:
Chun Xu and Fuqiang Zhang
Department of Prosthodontics, Shanghai Ninth
People’s Hospital
Shanghai Jiao Tong University School of Medicine
639 Zhizaoju Road
Shanghai 200011
China
Tel.: +86 21 23271699x5691
Fax: +86 21 63162608
e-mails: imxuchun@163.com (C. Xu), fqzhang@vip.
163.com (F. Zhang).
Key words: bone implant interactions, histo-pathology, surgical techniques
Abstract
Objective: This meta-analysis was aimed at assessing whether immediate dental implant placement
into infected vs. non-infected sites produced different effects on implant failure risk and marginal
bone loss.
Material and methods: Relevant studies were identified by searching articles in PubMed, Web of
Knowledge, and the Cochrane Library through February 2015 and by reviewing the reference lists
of the retrieved articles. When an intervention led to dichotomous outcomes, the outcomes were
expressed as risk ratios, whereas continuous outcomes were expressed as mean differences in
millimeters; each had a 95% confidence interval. Study-specific estimates were combined using
fixed-effects models.
Results: A total of 1743 articles were identified following the search process. Seven studies were
finally included in the meta-analysis, which comprised a total of 1586 implants and 25 failures.
Compared to the immediate insertion of a dental implant into a non-infected site, the insertion of
an implant into an infected site showed 116% increase in the risk of implant failure, which had
borderline statistical significance (risk ratio = 2.16, 95% confidence interval: 0.97, 4.80, P = 0.058;
heterogeneity: I2 = 0.0%, Pheterogeneity = 0.997). With regard to marginal bone loss, we observed no
statistically significant difference between insertions into infected vs. non-infected sites (mean
difference = 0.04, 95% confidence interval: 0.09, 0.02, P = 0.173, heterogeneity: I2 = 0.0%,
Pheterogeneity = 0.765).
Conclusion: This meta-analysis suggests that immediately placing a dental implant into an infected
site may increase the risk of implant failure. Given the presence of uncontrolled confounders in the
studies that were assessed, the results should be interpreted with caution.

Traditionally, before placing implants, tooth
extraction sockets are left to heal for several
months or longer (Adell et al. 1981). How-
ever, due to the preservation of esthetics, the
maintenance of the alveolar walls, and a
reduction of treatment time, it is sometimes
advisable to place the implant into the
extraction socket immediately, without wait-
ing for healing (Lazzara 1989; Vanden
Bogaerde et al. 2005).
Recently, the immediate placement of an
implant into an infected socket has become a
very controversial topic. A large collection of
animal studies has shown there to be no sig-
†
These authors contributed equally to this work.
nificant difference between the success rates
Date: of immediately placing an implant into an
Accepted 8 October 2015
infected socket vs. a non-infected socket
To cite this article: (Novaes et al. 2003; Palmer 2012), although
Zhao D, Wu Y, Xu C, Zhang F. Immediate dental implant
placement into infected vs. non-infected sockets: a meta- controversy still exists in human studies.
analysis. Several clinical studies have indicated that
Clin. Oral Impl. Res. 27, 2016, 1290–1296
doi: 10.1111/clr.12739 immediate implantation into an infected
socket may increase the failure risk of the
dental implant in comparison with implants
placed into non-infected sockets (Schwartz-
Arad & Chaushu 1997; Lindeboom et al.
2006; Marconcini et al. 2013), whereas in
other studies, no difference in risk was
observed (Del Fabbro et al. 2009; Truninger
et al. 2011). Although there are three reviews
about this topic, they gave different conclu-
sions and suggestions. One suggested that
implants could be placed into sites with peri-
apical and periodontal infections (Waasdorp
et al. 2010), while the other two gave
reserved suggestions (Alvarez-Camino et al.
2013; Chrcanovic et al. 2015).
Despite that a large number of relevant
studies have been conducted thus far, the
results are still conflicting and inconclusive
as to whether the presence of periodontal or
periapical infection is associated with
immediate implant failure and whether it is

1290 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
 Zhao et al
Infection and immediate dental implant failure

advisable to heal the infected socket prior to than one study was conducted within the related to the implants and not the patients.
implant placement. Therefore, we conducted same study population, we selected either The estimates of relative effect were
a meta-analysis of published clinical studies the most recent report or the report with expressed as a risk ratio (RR) in the case of a
to quantitatively assess whether the immedi- the most applicable estimates to the meta- dichotomous outcome and as a mean differ-
ate placement of dental implants into analysis. ence (MD) in millimeters in the case of a
infected vs. non-infected sites changed the continuous outcome; both required a 95%
risk of implant failure and marginal bone Data extraction confidence interval (CI). Study-specific esti-
loss. We used a standardized protocol and report- mates were combined using either fixed or
ing form to extract the following data from random effects models, depending on
each publication: the first author, the year of whether a statistical heterogeneity was
Material and methods publication, the study design, the number of detected (Der Simonian & Laird 1986).
patients, the patients’ ages, the follow-up Heterogeneity among studies was assessed
This study followed the Preferred Reporting time, the type of pathology, the implant heal- with the Q and the I2 statistics, and the
Items for Systematic Reviews and Meta-Ana- ing time, the number of patients with results were defined as heterogeneous for a P
lyses protocols 2015 (PRISMA-P 2015) state- implant failures, and the marginal bone loss. value <0.10 or an I2 > 50% (Higgins &
ment guidelines (Moher et al. 2015). If possible, we contacted the authors for Thompson 2002). Small study effects (effects
missing data. due to studies with small sample size), such
Study selection as publication bias, were evaluated by visual
We performed a systematic literature search Statistical analysis inspection of a funnel plot, and formal test-
in PubMed, Web of Knowledge, and the The infection and implant failure rates were ing was conducted using Egger’s and Begg’s
Cochrane Library through February 2015 treated as the dichotomous measures; the tests (Begg & Mazumdar 1994; Egger et al.
using the following key words: (implant OR rate of marginal bone loss was treated as a 1997).
implants) AND (infected OR infectious OR continuous variable. The statistical units for Statistical analyses were conducted using
periapical OR periodontal) AND (immediate “implant failure” and “marginal bone loss” Stata version 11.0 (StataCorp LP, College
OR immediately). The identified publications
were reviewed for their relevance to the
research topic by two independent authors
(ZD and WYQ). We also manually searched
the reference lists of the identified studies
and the relevant reviews to identify addi-
tional studies.
In this meta-analysis, oral infection is
defined as a periapical or periodontal lesion
identified by radiographs and chart notes.
The primary outcome of interest was implant
failure; the secondary outcome of interest
was marginal bone loss. Implant failure crite-
ria included clinically detectable implant
mobility during follow-up evaluations, radio-
graphic evidence of peri-implant radiolu-
cency, and signs or symptoms of infection,
according to the criteria reported by Albrek-
tsson (Albrektsson et al.1986).
To be included in the meta-analysis, each
study had to meet the following eligibility
criteria: (i) the study type was a clinical
human study, either randomized or not, and
was either interventional or observational; (ii)
the study compared the risks of implant fail-
ure or marginal bone loss for implants that
were immediately placed into infected vs.
non-infected sockets; (iii) implant failure and
marginal bone loss were the outcomes of
interest; (iv) the study either provided the
usable estimates and the corresponding 95%
CIs or enough data to calculate these mea-
sures. Exclusion criteria included case
reports, technical reports, animal studies,
in vitro studies, finite element analysis (FEA)
studies, and reviews. In cases where more Fig. 1. Selection process.

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd 1291 | Clin. Oral Impl. Res. 27, 2016 / 1290–1296
Zhao et al
Infection and immediate dental implant failure
Station, TX, USA). Two-sided P values
<0.05 were considered statistically
significant.
Results
Literature search
The study selection process is summarized
in Fig. 1. The literature search identified a
total of 1743 records. After excluding 596
records with duplicated titles and abstracts,
1147 records were obtained. After the title
and abstract screening, 1128 records were
further excluded based on the prespecified
inclusion and exclusion criteria. Thus, 19
records were selected for full-text screen-
ing. Twelve articles did not meet the
inclusion criteria after reading the full
texts, of which eight did not have control
groups, two were updated studies, one was
a duplicated study, and one was a review.
Finally, a total of seven publications were
included in the meta-analysis (Crespi et al.
2010a,b; Bell et al. 2011; Fugazzotto 2012;
Blus et al. 2015; Jung et al. 2013; Montoya-
Salazar et al. 2014). Additional manual
searching of the reference lists of
selected studies did not yield additional
publications.
Description of the studies
The descriptive details of the seven included
studies are listed in Table 1. There were five
prospective and two retrospective studies. A
total of 935 patients with 1586 implants
were included in the meta-analysis. All stud-
ies provided information on the patients’
ages, and none of them included non-adult
patients. Two studies included several
patients who smoked (Bell et al. 2011; Jung
et al. 2013). The implants were immediately
placed into infected sites after either being
irradiated with a yttrium–scandium–gallium–
garnet (Er,Cr:YSGG) laser (Montoya-Salazar
et al. 2014) or ultrasonicated (Fugazzotto
2012), which is a different method than is
used in other studies. In two of the studies,
the implants were from Straumann (Bell
et al. 2011; Jung et al. 2013), and two studies
did not indicate which brands of implants
were used (Fugazzotto 2012; Blus et al.
2015). All of the implants were inserted into
fresh extraction sockets, and information
was provided with regard to the healing/load-
ing time. In Crespi’s study (Crespi et al.
2010b), implants were immediately loaded,
and in Blus’ study (Blus et al. 2015),
implants were loaded either immediately,
early or delayed. All of the included studies
1292 | Clin. Oral Impl. Res. 27, 2016 / 1290–1296
had relatively long follow-up times (at least
12 months).
Of the seven studies, a total of 674 dental
implants were placed into infected sites, with
15 failures (2.23%); a total of 912 implants
were placed into non-infected sites, with 10
failures (1.10%). No individual study showed
a statistically significant difference concern-
ing the risks of implant failure between the
infected and non-infected (control) groups. In
two of the included studies, there were no
implant failures (Crespi et al. 2010b; Jung
et al. 2013). Three of the studies did not pro-
vide information about marginal bone levels
(Bell et al. 2011; Fugazzotto 2012; Blus et al.
2015); of the four studies providing this infor-
mation, only one study reported a change of
vertical distance from the implant shoulder
to the first bone-to-implant contact (IS-BIC)
on the mesial or the distal side of the
implant (Jung et al. 2013). Three studies
reported marginal bone loss during the fol-
low-up period (Crespi et al. 2010a,b; Mon-
toya-Salazar et al. 2014).
Quality assessment
Each trial was assessed for risk of bias
(Table 2). The trials were assessed in follow-
ing aspects: random sequence generation,
allocation concealment, blinding of outcome
assessments, and incomplete outcome data.
If all criteria were met, the trail would be
judged as low risk of bias. In each of the
studies, sequence generation was not ran-
domized, allocation concealment was inade-
quate, and there was no blinding. Thus, all
studies were judged to be at high risk of bias.
Meta-analysis
Compared with the insertion of a dental
implant into a non-infected site, inserting an
implant into an infected site doubled the risk
of implant failure, which had borderline
statistical significance (pooled RR = 2.16,
95% CI: 0.97, 4.80, P = 0.058; heterogene-
ity: I2 = 0.0%, Pheterogeneity = 0.997; Fig. 2).
When the studies that evaluated implants
with a standard time of loading were
pooled, a RR of 2.03 was the result (95%
CI: 0.88, 4.69, P = 0.10; heterogeneity:
I2 = 0%, Pheterogeneity = 0.945), whereas when
studies that evaluated implants with imme-
diate loading were pooled, a RR of 1.26 was
the result (95% CI: 0.16, 9.82, P = 0.82;
heterogeneity: I2 = 0%, Pheterogeneity = 0.687).
When studies that evaluated sites with peri-
apical lesions were pooled, a RR of 2.16 was
the result (95% CI: 0.90, 5.21, P = 0.09;
heterogeneity: I2 = 0%, Pheterogeneity = 0.922),
whereas when studies that evaluated sites
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
with periodontal lesions were pooled, a RR of
1.99 was the result (95% CI: 0.10, 41.09,
P = 0.66). And a subgroup analysis stratified
by different follow-up periods showed that
the pooled RRs were 2.05 (95% CI: 0.84,
4.99) and 2.58(95% CI: 0.43, 15.47) for the
<3-years and ≥3-years follow-ups, respec-
tively. No evidence of publication bias was
indicated by visual examination of the funnel
plot or the Egger’s (P = 0.252) and Begg’s tests
(P = 0.806) (Fig. 3).
For the secondary outcome, i.e., marginal
bone loss, there was no statistically signifi-
cant difference between infected and non-
infected sites, and the pooled MDoverall was
0.04 (95% CI: 0.09, 0.02, P = 0.17; hetero-
geneity: I2 = 0.00%, Pheterogeneity = 0.765).
The pooled MDs were 0.05 (95% CI: 0.15,
0.04, P = 0.25; heterogeneity: I2 = 0.0%,
Pheterogeneity = 0.164) and 0.03 (95% CI:
0.10, 0.04, P = 0.40; heterogeneity: I2 =
0.00%, Pheterogeneity = 0.490) for the 1-year
and ≥2 year follow-ups, respectively (Fig. 4).
Discussion
To our knowledge, this is the first meta-ana-
lysis to summarize the current evidence on
the association between the immediate
placement of an implant into an infected
socket and the implant failure risk. We
found a borderline significant difference on
the risks of implant failure between infected
and non-infected sites. The findings from
this meta-analysis suggested that the imme-
diate placement of implants into infected
sockets might increase the risk of implant
failure.
Currently, many animal studies have
shown that there are no significant differ-
ences between placing implants into infected
vs. non-infected sites with regard to either
implant failure or to changes in marginal
bone levels (Marcaccini et al. 2003; Novaes
et al. 2003; Papalexiou et al. 2004). These
results obtained in the animal studies may
be due to small sample sizes, short follow-up
times, or the creation of artificial periapical
or periodontal lesions that may not be repre-
sentative of human lesions. Several clinical
trials have suggested that implant placement
into sites with periodontal or endodontic
infections may increase implant failure risk
and/or marginal bone loss (Karoussis et al.
2003; Casap et al. 2007). This increased
failure risk may result from a negative
effect imparted by periapical and periodon-
tal infected lesions on osseointegration
(Werbitt & Goldberg 1992). Implants can also
 Table 1. Characteristics of studies included in meta-analysis
Follow-up, Failed/Placed Failure Antibiotics/Mouth Healing Marginal bone loss Type of
Study Design Patients, n Age, years year/month Implants, n Rate, % Rinse, days Period, m (Mean  SD), mm Infection Implant type
Blus et al. (2015) Prospective 168 (36,Ta; 47, 26–77 1 2/36 (Ta), 0/47 97.6 (T), Amoxicillin with Immediate; NM Acute, NM
Tc; 85,C) (Tc), 1/85 (C) 98.8 (C) Clavulanic acid, 1–3 m; 3 m Chronic
2 9 1 g/day for periapical
5 days, starting lesion
6–12 h before
extraction
Montoya-Salazar Prospective 36 (18,T; 18,C) 18–50 3 1/1 8 (T), 94.44z(T), Amoxicillin, 4.5 m 0.53  0.13 (T); Chronic Mis Ibe
rica,
et al. (2014) 0/1 8 (C) 100(C) 10 days (pre- 0.60  0.16 (C) periapical Barcelona,
operation lesion Spain
4 days)
Jung et al. (2013) Prospective 27 (12,T; 15,C) 53 (31–87), T; 5 0/12 (T), 100 (T), Amoxicyllin 3m Change IS- Periapical Standard Plus or
60 (28–82), C 0/15 (C) 100 (C) 750 mg, 5 days; BIC mesial: lesion Tapered Effect;
0.2% 0.2  0.4 (T); Straumann
Chlorhexidine 0.1  0.4 (C) Dental Implant
digluconate Change IS- System, Straumann
BIC distal: AG, Basel,
0.3  0.7 (T); Switzerland
0.2  0.6 (C)

© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Fugazzotto (2012) Retrospective 128 (64,T; 64, 46 (21–71) 62–64 m 3/64 (T), 95.3 (T), Amoxicillin 3–7 m NM Periapical NM
C) 1/64 (C) 98.4 (C) (39500 mg/day, pathology
for 10 days), or
Clindamycin
(29300 mg/day,
for 10 days).
Etodolac
49400 mg /day,
for 5 days
Bell et al. (2011) Retrospective 922 (285,T; 58.4 (T); 60.1 19.75 m 7/285 (T), 97.5 (T), Pre-operative 3m NM Chronic Strauman Dental
637,C) (C) (3–93 m) 8/637 (C) 98.7 (C) Chlorhexidine periapical Implants, Basel,
rinse, lesion Switzerland
intravenous
antibiotics
(600 mg
Clindamycin or
2 g Ampicillin)
Crespi et al. Prospective 275 (197,T; 78, 52.5 (32–71) 4 2/197(T), 98.9 (T), Amoxicillin Immediate 0.79  0.38 (T); Chronic Titanium Plasma

1293 |
(2010a) C) 0/78 (C) 100 (C) (291 g /day, for 0.78  0.38 (C) periodontal Spray, Sweden-
7 days) from 1 h lesion Martina, Padova,
before surgery Italy.
Chlorhexidine
twice daily for
15 days
Crespi et al. Prospective 30 (15,T; 15,C) 51.2 (34–71) 1 0/15 (T), 100 (T), Amoxicillin 3m 0.86, 0.54 (T); Chronic Seven, Sweden-
(2010b) 0/15 (C) 100 (C) (291 g /day, for 0.82,0.52 (C) periapical Martina, Padova,
7 days) from 1 h lesion Italy.
before surgery
Chlorhexidine
twice daily for
15 days

T, the text (infected) group; C, the control (non-infected) group; Ta, the acute infected group; Tc, the chronic infected group; NM, no mention.
Zhao et al
Infection and immediate dental implant failure

Clin. Oral Impl. Res. 27, 2016 / 1290–1296

Zhao et al
Infection and immediate dental implant failure
Table 2. Risk of bias analysis dures to perform the decontamination of the
Incomplete outcome implant site, such as preoperative antibiotics,
Study data addressed postoperative antibiotics, and alveolar
Blus et al. (2015) No debridement, have been suggested to create
Montoya-Salazar No the appropriate conditions for bone regenera-
et al. (2014)
tion and osseointegration (Lindeboom et al.
Jung et al. (2013) Yes
Fugazzotto (2012) No 2006; Siegenthaler et al. 2007).
Bell et al. (2011) No There are several important issues in the
Crespi et al. (2010a) No present meta-analysis that should be taken
Crespi et al. (2010b) No
into consideration. First, several of the
For all studies, sequence generation was not included publications had only short-term
randomized; allocation concealment was
follow-up periods, which may explain why
inadequate; there was no blinding; and the
estimated potential risk of bias was high. the failure risks were low in some studies
and why some studies reported no failure
events at all. A longer follow-up period can
potentially become contaminated due to rem- lead to an increase in the risk of failure.
nants of infection, which may also lead to Because of the short follow-up periods in the
implant failure (Polizzi et al. 2000). In cur- included studies, it was difficult to discern
rent clinical practice, proper clinical proce- statistically significant differences on the
Fig. 2. Forest plot for implant failures regarding infected group vs. non-infected group.
Fig. 3. Forest plot of margin bone levels for 1 year and no less than 2 years.
1294 | Clin. Oral Impl. Res. 27, 2016 / 1290–1296 © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
single study level with respect to the risk of
implant failure between infected groups and
non-infected control groups. However, a
meta-analysis of these individual studies
could help gain a higher statistical power for
the measure of interest, as opposed to a less
precise measure derived from a single study.
Second, only one of the included studies was
about periodontal infection. Despite being
different as to etiology and pathogenesis,
periodontal infections and periapical infec-
tions show some similarities. They share a
common microbiota (Sundqvist 1992; Noiri
et al. 2001) and both elevate systemic cyto-
kine levels (Barkhordar et al. 1999; Gamonal
et al. 2000). The similarity suggests that
cross-infection between the root canal and
the periodontal pocket can occur. When we
excluded the study about periodontal infec-
tion, the results were basically the same;
however, the generalization of these results
should be taken with caution. Third, in two
included studies, two implants (one was for
test and the other for control subject) placed
in the same patient. When we excluded
these two studies from the analysis, the
pooled results of the remaining studies did
not change materially. Forth, as follow-up
time increased, the differences in marginal
bone loss between the intervention and the
control groups decreased, which suggested
that marginal bone loss primarily occurs
within the first year following implant place-
ment (Jung et al. 2013). With the passage of
time, the autoimmunization of the patient
may defeat the infection, which results in a
phase of stability (i.e., the marginal bone
level no longer substantially changes) (Block
et al. 2009).
Our study had several strengths. The pre-
sent article is the first meta-analysis to quan-
titatively assess the association between the
immediate placement of a dental implant
into an infected socket and the implant fail-
ure risk. Moreover, no significant heterogene-
ity was detected between the included
studies. However, this study also had limita-
tions. First, two of the included studies had a
retrospective design, and the nature of a ret-
rospective study is inherently flawed. Second,
several unknown or unmeasured con-
founders, such as smoking, diabetes, the
pathological extent of the infections, and
brushing habits, may have affected implant
failure risks or marginal bone loss. However,
it is difficult to avoid the residual confound-
ing that was inherent to the original studies
when using a meta-analyzed approach. Third,
one of the included studies used different
procedures (i.e., the application of a laser to

Fig. 4. Funnel plot for implant failures regarding infected group vs. non-infected group.
the infected site) between the test and con-
trol groups, which might have affected the
failure risk. However, when we excluded this
study from the analysis, the pooled results
References
Adell, R., Lekholm, U., Rockler, B. & Branemark,
P.I. (1981) A 15-year study of osseointegrated
implants in the treatment of the edentulous jaw.
International Journal of Oral Surgery 10: 387–
416.
Albrektsson, T., Zarb, G., Worthington, P. & Eriks-
son, A.R. (1986) The long-term efficacy of cur-
rently used dental implants: a review and
proposed criteria of success. The International
Journal of Oral & Maxillofacial Implants 1:
11–25.
Alvarez-Camino, J.C., Valmaseda-Castellon, E. &
Gay-Escoda, C. (2013) Immediate implants placed
in fresh sockets associated to periapical infec-
tious processes. A systematic review. Medicina
oral, patologia oral y cirugia bucal 18: e780–
e785.
Barkhordar, R.A., Hayashi, C. & Hussain, M.Z.
(1999) Detection of interleukin-6 in human den-
tal pulp and periapical lesion. Endodontics &
Dental Traumotology 15: 26–27.
Begg, C.B. & Mazumdar, M. (1994) Operating char-
acteristics of a rank correlation test for publica-
tion bias. Biometrics 50: 1088–1101.
Bell, C.L., Diehl, D., Bell, B.M. & Bell, R.E. (2011)
The immediate placement of dental implants into
extraction sites with periapical lesions: a retro-
spective chart review. Journal of Oral and Max-
illofacial Surgery 69: 1623–1627.
Block, M.S., Mercante, D.E., Lirette, D., Mohamed,
W., Ryser, M. & Castellon, P. (2009) Prospective
evaluation of immediate and delayed provisional
single tooth restorations. Journal of Oral and
Maxillofacial Surgery 67: 89–107.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Zhao et al
Infection and immediate dental implant failure
did not change. Finally, publication bias
might have influenced the results. Although
there is no proof of publication bias in the
present meta-analysis, tests for publication
Blus, C., Szmukler-Moncler, S., Khoury, P. & Orru,
G. (2015) Immediate implants placed in infected
and noninfected sites after atraumatic tooth
extraction and placement with ultrasonic bone
surgery. Clinical Implant Dentistry and Related
Research 17(Suppl. 1): e287–e297.
Casap, N., Zeltser, C., Wexler, A., Tarazi, E. &
Zeltser, R. (2007) Immediate placement of dental
implants into debrided infected dentoalveolar
sockets. Journal of Oral and Maxillofacial
Surgery 65: 384–392.
Chrcanovic, B.R., Martins, M.D. & Wennerberg, A.
(2015) Immediate placement of implants into
infected sites: a systematic review. Clinical
Implant Dentistry and Related Research 17
(Suppl. 1): e1–e16.
Crespi, R., Cappare, P. & Gherlone, E. (2010a)
Immediate loading of dental implants placed in
periodontally infected and non-infected sites: a 4-
year follow-up clinical study. Journal of Periodon-
tology 81: 1140–1146.
Crespi, R., Cappare, P. & Gherlone, E. (2010b) Fresh-
socket implants in periapical infected sites in
humans. Journal of Periodontology 81: 378–383.
Del Fabbro, M., Boggian, C. & Taschieri, S. (2009)
Immediate implant placement into fresh extrac-
tion sites with chronic periapical pathologic fea-
tures combined with plasma rich in growth
factors: preliminary results of single-cohort study.
Journal of Oral and Maxillofacial Surgery 67:
2476–2484.
Der Simonian, R. & Laird, N. (1986) Meta-analysis
in clinical trials. Controlled Clinical Trials 7:
177–188.
1295 |
bias have low statistical power, because of
the limited number of studies.
Conclusion
This study suggests that the immediate
placement of a dental implant into an
infected site might increase the failure risk
of the implant in comparison with an
implant that was placed into a non-infected
site. The results of the present meta-analysis
should be interpreted with caution because of
the presence of uncontrolled confounding fac-
tors in the included studies. Moving forward,
despite the limitations of this study, clinical
practitioners should consider the possibly
increased failure risk that is associated with
the immediate placement of dental implants
into infected sockets.
Acknowledgements: The authors
declare no funding for the study and no
conflict of interest. We appreciate Dr. Yang
Yang for his contribution to the statistical
analysis.
Egger, M., Davey Smith, G., Schneider, M. & Min-
der, C. (1997) Bias in meta-analysis detected by a
simple, graphical test. BMJ 315: 629–634.
Fugazzotto, P.A. (2012) A retrospective analysis of
implants immediately placed in sites with and
without periapical pathology in sixty-four
patients. Journal of Periodontology 83: 182–186.
Gamonal, J., Acevedo, A., Bascones, A., Jorge, O. &
Silva, A. (2000) Levels of interleukin-1 beta, -8,
and -10 and Rantes in gingival crevicular fluid
and cell population in adult periodontitis patients
and the effect of periodontal treatment. Journal of
Periodontology 71: 1535–1545.
Higgins, J.P. & Thompson, S.G. (2002) Quantifying
heterogeneity in a meta-analysis. Statistics in
Medicine 21: 1539–1558.
Jung, R.E., Zaugg, B., Philipp, A.O., Truninger, T.C.,
Siegenthaler, D.W. & Hammerle, C.H. (2013) A
prospective, controlled clinical trial evaluating
the clinical radiological and aesthetic outcome
after 5 years of immediately placed implants in
sockets exhibiting periapical pathology. Clinical
Oral Implants Research 24: 839–846.
Karoussis, I.K., Salvi, G.E., Heitz-Mayfield, L.J.,
Bragger, U., Hammerle, C.H. & Lang, N.P. (2003)
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
14: 329–339.
Lazzara, R.J. (1989) Immediate implant placement
into extraction sites: surgical and restorative
advantages. The International Journal of Peri-
odontics & Restorative Dentistry 9: 332–343.
Clin. Oral Impl. Res. 27, 2016 / 1290–1296
Zhao et al
Infection and immediate dental implant failure
Lindeboom, J.A., Tjiook, Y. & Kroon, F.H. (2006)
Immediate placement of implants in periapical
infected sites: a prospective randomized study in
50 patients. Oral Surgery, Oral Medicine, Oral
Pathology, Oral Radiology, and Endodontics 101:
705–710.
Marcaccini, A.M., Novaes, A.B., Jr, Souza, S.L.,
Taba, M., Jr & Grisi, M.F. (2003) Immediate
placement of implants into periodontally infected
sites in dogs. Part 2: A fluorescence microscopy
study. The International Journal of Oral & Max-
illofacial Implants 18: 812–819.
Marconcini, S., Barone, A., Gelpi, F., Briguglio, F. &
Covani, U. (2013) Immediate implant placement
in infected sites: a case series. Journal of peri-
odontology. 84: 196–202.
Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Lib-
erati, A., Petticrew, M., Shekelle, P. & Stewart,
L.A. (2015) Preferred reporting items for systematic
review and meta-analysis protocols (PRISMA-P)
2015 statement. Systematic Reviews 4: 1.
Montoya-Salazar, V., Castillo-Oyague, R., Torres-
Sanchez, C., Lynch, C.D., Gutierrez-Perez, J.L. &
Torres-Lagares, D. (2014) Outcome of single
immediate implants placed in post-extraction
infected and non-infected sites, restored with
cemented crowns: a 3-year prospective study.
Journal of Dentistry 42: 645–652.
Noiri, Y., Li, L. & Ebisu, S. (2001) The localization
of periodontal-disease associated bacteria in
1296 | Clin. Oral Impl. Res. 27, 2016 / 1290–1296
human periodontal pockets. Journal of Dental
Research 80: 1930–1934.
Novaes, A.B., Jr, Marcaccini, A.M., Souza, S.L.,
Taba, M., Jr & Grisi, M.F. (2003) Immediate
placement of implants into periodontally
infected sites in dogs: a histomorphometric
study of bone-implant contact. The Interna-
tional Journal of Oral & Maxillofacial Implants
18: 391–398.
Palmer, R. (2012) Evidence for survival of implants
placed into infected sites is limited. The Journal
of Evidence-Based Dental Practice 12: 187–188.
Papalexiou, V., Novaes, A.B., Jr, Grisi, M.F., Souza,
S.S., Taba, M., Jr & Kajiwara, J.K. (2004) Influence
of implant microstructure on the dynamics of
bone healing around immediate implants placed
into periodontally infected sites. A confocal laser
scanning microscopic study. Clinical Oral
Implants Research 15: 44–53.
Polizzi, G., Grunder, U., Goene, R., Hatano, N.,
Henry, P., Jackson, W.J., Kawamura, K., Renouard,
F., Rosenberg, R., Triplett, G., Werbitt, M. & Lith-
ner, B., (2000) Immediate and delayed implant
placement into extraction sockets: a 5-year report.
Clinical Implant Dentistry and Related Research
2: 93–99.
Schwartz-Arad, D. & Chaushu, G. (1997) The ways
and wherefores of immediate placement of
implants into fresh extraction sites: a literature
review. Journal of Periodontology 68: 915–923.
© 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Siegenthaler, D.W., Jung, R.E., Holderegger, C.,
Roos, M. & Hammerle, C.H. (2007) Replacement
of teeth exhibiting periapical pathology by imme-
diate implants: a prospective, controlled clinical
trial. Clinical Oral Implants Research 18: 727–
737.
Sundqvist, G. (1992) Ecology of the root canal flora.
Journal of Endodontics 18: 427–430.
Truninger, T.C., Philipp, A.O., Siegenthaler, D.W.,
Roos, M., Hammerle, C.H. & Jung, R.E. (2011) A
prospective, controlled clinical trial evaluating
the clinical and radiological outcome after 3 years
of immediately placed implants in sockets
exhibiting periapical pathology. Clinical oral
implants research 22: 20–27.
Vanden Bogaerde, L., Rangert, B. & Wendelhag, I.
(2005) Immediate/early function of Branemark
System TiUnite implants in fresh extraction
sockets in maxillae and posterior mandibles: an
18-month prospective clinical study. Clinical
Implant Dentistry and Related Research 7(Suppl.
1): S121–S130.
Waasdorp, J.A., Evian, C.I. & Mandracchia, M.
(2010) Immediate placement of implants into
infected sites: a systematic review of the litera-
ture. Journal of Periodontology 81: 801–808.
Werbitt, M.J. & Goldberg, P.V. (1992) The immedi-
ate implant: bone preservation and bone regenera-
tion. The International Journal of Periodontics &
Restorative Dentistry 12: 206–217.