Analysis of Bacterial Flora Associated with

Peri-implantitis Using Obligate Anaerobic

Culture Technique and 16S rDNA Gene Sequence
Naoki Tamura, DDS, PhD1/Morio Ochi, DDS, PhD2/Hiroshi Miyakawa, PhD3/Futoshi Nakazawa, PhD4
Purpose: To analyze and characterize the predominant bacterial flora associated with peri-implantitis by
using culture techniques under obligate anaerobic conditions and 16S rDNA gene sequences. Materials
and Methods: Subgingival bacterial specimens were taken from 30 patients: control (n = 15), consisting
of patients with only healthy implants; and test (n = 15), consisting of patients with peri-implantitis. In
both groups, subgingival bacterial specimens were taken from the deepest sites. An anaerobic glove box
system was used to cultivate bacterial strains. The bacterial strains were identified by 16S rDNA genebased polymerase chain reaction and comparison of the gene sequences. Results: Peri-implantitis sites
had approximately 10-fold higher mean colony forming units (per milliliter) than healthy implant sites.
A total of 69 different bacterial species were identified in the peri-implantitis sites and 53 in the healthy
implant sites. The predominant bacterial species in the peri-implantitis sites were Eubacterium nodatum,
E brachy, E saphenum, Filifactor alocis, Slackia exigua, Parascardovia denticolens, Prevotella intermedia,
Fusobacterium nucleatum, Porphyromonas gingivalis, Centipeda periodontii, and Parvimonas micra.
The predominant bacteria in healthy implant sites apart from Streptococcus were Pseudoramibacter
alactolyticus, Veillonella species, Actinomyces israelii, Actinomyces species, Propionibacterium acnes,
and Parvimonas micra. Conclusion: These results suggest that the environment in the depths of the
sulcus showing peri-implantitis is well suited for growth of obligate anaerobic bacteria. The present study
demonstrated that the sulcus around oral implants with peri-implantitis harbors high levels of asaccharolytic
anaerobic gram-positive rods (AAGPRs) such as E nodatum, E brachy, E saphenum, Filifactor alocis, Slackia
exigua, and gram-negative anaerobic rods, suggesting that conventional periodontopathic bacteria are not
the only periodontal pathogens active in peri-implantitis, and that AAGPRs may also play an important role.
Int J Oral Maxiilofac Implants 2013;28:15211529. doi: 10.11607/jomi.2570
Key words: 16S rDNA, bacterial flora, obligate anaerobe, peri-implantitis

1 Assistant

Professor, Division of Fixed Prosthodontics and Oral

Implantology, Department of Oral Rehabilitation, School of
Dentistry, Health Sciences University of Hokkaido, Kanazawa,
Hokkaido, Japan.
2Professor and Chairman, Division of Fixed Prosthodontics and
Oral Implantology, Department of Oral Rehabilitation, Health
Sciences University of Hokkaido, Kanazawa, Hokkaido, Japan.
3Assistant Professor, Department of Oral Microbiology,
School of Dentistry, Health Sciences University of Hokkaido,
Kanazawa, Hokkaido, Japan.
4Professor and Chairman, Department of Oral Microbiology,
School of Dentistry, Health Sciences University of Hokkaido,
Kanazawa, Hokkaido, Japan.
Correspondence to: Dr Naoki Tamura, Division of Fixed
Prosthodontics and Oral Implantology, Department of Oral
Rehabilitation, School of Dentistry, Health Sciences University
of Hokkaido; 1757 Kanazawa Ishikari-Tobetsu Hokkaido,
061-0293, Japan. Fax: +81-133-23-2892.
Email: naokikid@hoku-iryo-u.ac.jp
2013 by Quintessence Publishing Co Inc.

sseointegrated oral implants are a major tool in

prosthetic dentistry. Although most implants are
successful, implant failures occasionally occur. Failing
implants are frequently characterized by loss of supporting bone, and these implants must be removed.
The reasons for failure of implants are multiple (occlusion, parafunction, overloading, premature loading,
and bacterial infection).1 Above all, two principal reasons for implant failure are mechanical stress and/or
bacterial infection. Implants that fail after demonstrating osseointegration are considered to be late failures.
The factors contributing to late or delayed implant failure such as peri-implantitis are not well-known, and
many questions about etiology and treatment remain
unanswered. Peri-implantitis is defined as an irreversible inflammatory reaction associated with loss of supporting bone around an osseointegrated implant in
function.2,3 Two cross-sectional studies reported that
peri-implantitis was identified in approximately 28%

The International Journal of Oral & Maxillofacial Implants 1521

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

and 56% of subjects and in 12% and 43% of implant

sites, respectively.3 Signs of a failing dental implant are
detected both clinically and radiographically, with the
diagnosis made in a similar way to periodontitis.4
The initially sterile and clean surface of an implant
enters the totally different environment of adult gingival
or periodontal tissues and bacterial flora, offering a new
surface in the oral cavity for adherence and colonization of bacteria. Biofilm formation on oral implants can
cause inflammation of peri-implant tissue, endangering
the long-term success of osseointegrated implants.5 It is
known that a large diversity of bacterial species inhabit
the peri-implant area. A common finding in previous
studies was that similar bacterial flora was found around
oral implants and natural teeth.6 A high proportion of
coccoid cells and facultative bacterial species and low
frequencies of periodontal pathogens including gramnegative bacteria were detected in healthy implant
sites. Clinical observations have recently revealed that
early colonization patterns differ between implant and
tooth surfaces.7 On the other hand, many reports have
indicated that the infected lesions in the sulcus around
peri-implantitis sites are traditionally associated with
chronic periodontopathic bacteria.8,9 It has traditionally been thought that peri-implantitis is associated
with a group of obligate anaerobic gram-negative rods
(OGNRs), in particular black-pigmented and motile
rods, and that gram-negative anaerobic cocci such as
Veillonella also play an important role.1,6,10 However,
these periodontopathic bacteria have also been observed in healthy implant sites.11
It has long been known that there are many uncultured, undescribed, and unknown bacterial species in
the human oral cavity.1214 Some oral bacteria, such as
asaccharolytic anaerobic gram-positive rods (AAGPRs),
are fastidious and grow poorly on artificial culture media; therefore, some of them are still not well known.
Numerous bacterial strains of AAGPRs have been isolated from human oral infectious lesions, several of
which have been documented as etiologic agents of
chronic periodontitis15 based on their frequent isolation from diseased periodontal sites.
Previously, new isolates of AAGPRs were classified as
members of the genus Eubacterium, which has historically acted as a repository for a large number of diverse
organisms16 and contains species and groups that are
phylogenetically unrelated.12 Recent studies have described many oral AAGPR isolates, some of which qualify as members of the genus Eubacterium but could
not be assigned to any of the established species. In
addition, several novel genera have been proposed
for some of the AAGPRs, eg, Pseudoramibacter, Slackia,
Filifactor, and Mogibacterium, and some species of
Eubacterium have been transferred to the novel genera
according to their phylogenetic characteristics. It has

been shown that antibody titers against some species

of AAGPRs in the sera of patients with periodontitis
are higher than those of healthy people, suggesting
that these bacteria cause immunologic reactions in
periodontal lesions.1,2,17 Nevertheless, AAGPRs are
not well-known because of the difficulty in cultivating them, the small size of the colony, and their lack
of reactivity in conventional biochemical tests. Thus,
further studies are needed to better understand the
asaccharolytic microorganisms under the obligate anaerobic conditions of peri-implantitis.
Recently, the analysis of 16S rDNA gene-based
polymerase chain reaction (PCR) has become a useful
approach to assess the phylogenetic and taxonomic
diversity of bacterial isolates.15,16,18 Culturing under
obligate anaerobic conditions is also an important
technique to allow the detection of almost every species in a given sample, revealing the presence of previously uncultivated and unclassified strains.
Peri-implant infection by pathogenic bacteria cannot be understood without a comprehensive knowledge of obligate anaerobes. Therefore, it seems
appropriate to compare the bacterial flora of periimplantitis with that of healthy implants by using culture techniques under obligate anaerobic conditions
and 16S rDNA gene-based PCR. The aim of this study
was to identify and characterize the predominant anaerobic bacterial flora in the sulcus around progressive
peri-implantitis in partially edentulous patients.

Materials and Methods

Patient and Sample Collection

This study was approved by the ethics committee

of the Health Sciences University of Hokkaido and
the Health Sciences University of Hokkaido Hospital
(Kanazawa, Hokkaido, Japan). The examination was
performed with the understanding and written consent of each patient. The patients were categorized into
two groups: the peri-implantitis group and the healthy
implant group. Medical and dental histories were obtained, and full-mouth periodontal and implant examinations were performed. Both groups had one or more
dental implants exposed for more than 6 months to the
oral environment. The patients in the peri-implantitis
group had clinical signs of peri-implantitis around one
or more implants. The patients in the healthy implant
group had no clinical signs of peri-implantitis. One implant was selected from each patient. Patients were excluded if they had severe systemic disease, had taken
antibacterial medication, had used mouth rinses during the previous 6 weeks, or had received peri-implant
therapy within the previous 6 months that could interfere with the clinical parameters evaluated.

1522 Volume 28, Number 6, 2013

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

Peri-implantitis sites were characterized by a probing depth (PD) of 4 mm, suppuration, bleeding on
probing (BOP),19 and visible three-thread loss of alveolar bone clearly extending around the implant as
visualized on radiographs.3,20 PD was measured in the
deepest pocket of each quadrant between the gingival
margin and the base of the peri-implant pocket with a
pressure sensitive probe (Click-Probe, KerrHawe).21 BOP
was recorded as present or absent. Radiographic bone
loss around implants was evaluated from available radiographs (periapical or orthopantomography) from
the fixture-abutment junction or the shoulder of the
implants, or the cemento-enamel junction (Table 1).
Based on these data, the diagnosis of peri-implantitis
was made.

Microbial Sampling

Sample sites were isolated with cotton rolls, air dried,

and supragingival plaque was carefully removed using
a pledget of cotton wool to avoid contamination. Subgingival bacterial specimens were taken with sterile
endodontic paper points (ISO #50), which were gently
inserted to the deepest point of each peri-implant or
periodontal lesion and kept in place for 30 seconds.
The paper points were then removed and placed in
an anaerobic sterile transport vial containing 1 mL of
semisolid medium (redox potential: 200 to approximately 400 mV, 3.7% brain heart infusion, 0.2% agar).
Samples were transferred as soon as possible (within
30 minutes) to an anaerobic glove box containing 80%
nitrogen, 10% hydrogen, and 10% carbon dioxide.

Microbiological Analysis

For in vitro evaluation, the suspension, dispersion, and

dilution of samples and the culture of bacterial strains
in the samples were carried out in the anaerobic
glove box. The inside of the anaerobic glove box was
maintained under 300 mV in redox potential during
all experimental procedures. All plates, media, buffer
solutions, and experimental instruments were kept in
the anaerobic glove box for at least 7 days before use.
Anaerobic bacterial isolation was executed by the following standard procedures.
Each sample was suspended in 1 mL of sterilized
phosphate-buffered saline (PBS) and mechanically dispersed with a teflon homogenizer and vortex mixer for
30 seconds. Serial 10-fold dilutions (0.1 mL each, from
102 to 106) were spread onto the surface of nonselective blood medium (BHI blood agar plate; 3.7%
brain heart infusion, 1.5% agar, 5% defibrinated sheep
blood, 0.1% hemin, 0.1% menadione) and incubated
in the anaerobic glove box at 37C for 7 days. Bacterial
strains found on the nonselective blood medium were
enumerated, and their percentage of the total number
of colony forming units (CFUs) was calculated. When

Table 1 Patient Selection Criteria

Peri-implantitis
Healthy
site
implant site
Bleeding on probing

Suppuration

Probing pocket depth (PD)*

4 mm

< 4 mm

Radiographic bone loss**

Periodontal maintenance care

Antibacterial agent
Time since implant placement

None

None

> 6 mo

> 6 mo

*Probing pocket depth assessed as greatest distance between

gingival margin and base of the peri-implant pocket.
**Radiographic bone loss around implants was evaluated from
available radiographs from the fixture-abutment junction or the
shoulder of the implants, or the cemento-enamel junction.

the number of colonies on a BHI blood agar plate did

not exceed 100, all of these colonies were isolated and
subcultured for identification.
Subcultured colonies were incubated anaerobically
or aerobically for 3 days, and for the purposes of this
study, obligate anaerobes were defined as bacteria that
grew only in the anaerobic glove box. In addition, these
subcultured isolates were examined by gram staining.

Primer Selection and 16S rDNA

Gene-Based PCR

Four primer sets were included in individual PCR assays.

The following universal primers were used for amplification of approximately 1,500 bp 16S rDNA22,23: forward
primer 16S27F (5-AGAGTTTGATCCTGGCTCAG-3),
16S341F (5-CCTACGGGAGGCAGCAG-3), reverse primer 16S1492R (5-TACGGCTACCTTGTTACGACTT-3), and
16S907R (5-CCGTCAATTCCTTTGAGTTT-3). Each DNA
amplification technique by colony-directed PCR was
performed in a total volume of 50 L in 0.2 mL microreaction tubes (MicroAmp, PE Biosystems), containing 1PCR buffer with 23 L deionized water, primers
at a concentration of 0.2 mol/L, and 25 L of DNA
polymerase (AmpliTaq Gold, PCR Master Mix, Applied
Biosystems).
The amplification program was as follows: preheating at 95C for 15 minutes, 35 cycles of denaturing
at 94C for 1 minute, annealing at 52C for 1 minute,
extension at 72C for 5 minutes, and a final extension step for 10 minutes at 72C using a thermocycler
(Veriti, Applied Biosystems). The amplification products were subjected to electrophoresis on 2.0% agarose gel and visualized by ethidium bromide staining.

Sequencing of 16S rDNA and

Phylogenetic Analysis

The PCR products obtained above were sequenced

at Takara Bio. Then, 16S rDNA sequences of bacterial
The International Journal of Oral & Maxillofacial Implants 1523

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

Fig 1Proportional distribution

of bacterial genera at peri-implantitis sites: orange, OGNRs;
blue, AAGPRs. Gram staining
characteristics: green, gram-positive cocci; purple, gram-negative
cocci; light blue, gram-positive
rods; red, gram-negative rods.
The bacterial flora of peri-implantitis sites consisted of bacteria
of 31 genera.

Table 2 Clinical Features of Patients

Clinical features

Peri-implantitis
site

Healthy implant
site

Patients

15

15

Age (y)

56.9
(range, 4676)

63.4
(range, 4677)

Sex (F/M)

8/7

4/11

Mean probing
depths

6.8 (max, 10 mm;

min, 4 mm)

1.3 (max, 3 mm;

min, 1 mm)

Mean CFUs/mL*

6.34 0.52**

5.16 0.86

CFUs: colony-forming units; max: maximum; min: minimum.

*Logarithm CFUs/mL
**Statistically different from healthy implant sites, P < .01,
Mann-Whitney U test.

strains were compared with 16S rDNA gene sequences

from the GenBank database using the BLAST search
program through the website of the DNA Data Bank of
Japan (DDBJ). A 16S rDNA gene sequence similarity of
98% was used as the cutoff for positive identification
of taxa.24,25

Statistical Analysis

The Mann-Whitney U test (P < .01: average total colonyforming units, P < .05: the proportional distribution of
OGNRs and AAGPRs) was used for statistical analyses
of the significant differences between the two groups.

Results
Clinical features of patients and sites selected for bacterial sampling are summarized in Table 2. A total of 30
partially edentulous patients involved in this study were
selected from systemically healthy patients referred to
the Health Sciences University of Hokkaido and Health
Sciences University of Hokkaido Hospital. The patients
included 12 females and 18 males. Fifteen patients
with peri-implantitis (8 females and 7 males) and 15 patients with healthy implants (4 females and 11 males)
were characterized as shown in Table 2. The implants of
the peri-implantitis patients included three types. One
implant was made of single crystal aluminum oxide,
11 were made of commercially pure titanium, and three
implants were coated with hydroxyapatite. The healthy
implants were made of commercially pure titanium.
The mean age was 56.9 years (range, 46 to 76 years) in
the peri-implantitis group and 63.4 years (range, 46 to
77 years) in the healthy implant group. The mean PD
and CFUs were higher in the peri-implantitis group
than in the healthy implant group. The mean PD was
6.8 mm (range, 4 to 10 mm) in the peri-implantitis sites
and 1.3 mm (range, 1 to 3 mm) in the healthy implant
sites. A high proportion of patients with a diagnosis of
peri-implantitis presented with a generalized loss of
supporting bone around the implants.
Bacterial strains were detected in all colonies cultured on the nonselective blood medium under anaerobic conditions by serial 10-fold dilutions to a
level 1104 per mL. Quantitative bacterial analysis
of peri-implantitis sites showed an average total of
6.34 0.52 colony-forming units (logarithm CFUs/mL).

1524 Volume 28, Number 6, 2013

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

Fig 2Proportional distribution

of bacterial genera at healthy
implant sites: orange, OGNRs;
blue, AAGPRs. Gram staining
characteristics: green, gram-positive cocci; purple, gram-negative
cocci; light blue, gram-positive
rods; red, gram-negative rods.
The bacterial flora of healthy implant sites consisted of bacteria
of 20 genera.

40
P < .05

30

20

10

Mean proportion of OGNRs (%)

Mean proportion of AAGPRs (%)

40

P < .05
30

20

10

0
Peri-implantitis

Healthy implants

Peri-implantitis

Healthy implants

Fig 3Mean proportional distribution of AAGPRs. The mean

proportional distribution was statistically significantly higher in
the peri-implantitis group (P < .05, Mann-Whitney U test).

Fig 4 Mean proportional distribution of OGNRs. The mean proportional distribution was statistically significantly higher in the
peri-implantitis group (P < .05, Mann-Whitney U test).

In contrast, the average total CFUs in the healthy implant group was 5.16 0.86. Significant differences
were seen between the CFUs of the peri-implantitis
and healthy implant groups (P < .01) (Table 2).
Figures 1 and 2 show the proportional distribution of bacterial genera in peri-implantitis sites and in
healthy implant sites, respectively. Streptococcus was
the main genera present at both healthy and periimplantitis sites. The flora observed at peri-implantitis
sites consisted of bacteria of 31 genera, but only 20
genera were detected at healthy implant sites (Figs 1
and 2). The total proportion of OGNRs and AAGPRs

accounted for about 40% of all CFUs at peri-implantitis

sites, in contrast to about 10% at healthy implant sites.
At peri-implantitis sites, the predominant genera were
Streptococcus (34%), Eubacterium (13%), Prevotella
(10%), Actinomyces (6%), and Fusobacterium (4%). At
healthy implant sites, the predominant genera were
Streptococcus (45%), Actinomyces (14%), Veillonella
(14%), and Propionibacterium (8%). The proportion
of Streptococcus at healthy implant sites was larger
than at peri-implantitis sites. At peri-implantitis sites,
Eubacterium that were AAGPRs and Prevotella that
were OGNRs were the next most predominant genera.
The International Journal of Oral & Maxillofacial Implants 1525

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

Fig 5 Proportional distribution

of the 69 different bacterial species identified at peri-implantitis sites: orange, OGNRs; blue,
AAGPRs.

Fig 6Proportional distribution

of the 53 different bacterial species identified at healthy implant
sites: orange, OGNRs; blue,
AAGPRs.

Figures 3 and 4 represent the mean proportional distribution of AAGPRs and OGNRs. AAGPRs were detected
in the peri-implantitis group (18%) and in the healthy
implant group (3%). OGNRs were detected in the periimplantitis group (20%) and in the healthy implant
group (6%). These data reveal statistically significant
differences between the two groups. The mean pro-

portional distributions were statistically significantly

higher in peri-implantitis (P < .05).
Peri-implantitis sites were colonized by complex
bacteria with a large proportion of OGNRs and AAGPRs.
A total of 69 different bacterial species were identified
in the peri-implantitis sites (Fig 5) and 53 in the healthy
implant sites (Fig 6).

1526 Volume 28, Number 6, 2013

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

Significant differences in bacterial species were

observed between healthy implant sites and periimplantitis sites. The predominant obligate anaerobic bacterial species at the peri-implantitis sites were
E nodatum (7%), P intermedia (5%), F nucleatum (3%),
Filifactor alocis (3%), E brachy (3%), Parascardovia
denticolens (3%), and Parvimonas micra (3%), while the
predominant obligate anaerobic bacterial species at
the healthy implant sites were Veillonella species (spp)
(14%), Propionibacterium acnes (5%), Pseudoramibacter
alactolyticus (3%), and Parvimonas micra (2%).

Discussion
This study aimed to identify the predominant bacteria
and characterize the anaerobic bacterial flora in the
sulcus around progressive peri-implantitis. Selection
of the patients for each group was based on clinical
conditions such as BOP, suppuration, PD, and radiographic criteria (Table 1).
In this study, various designs of implant were represented in the peri-implantitis group. Previous studies
have found no statistically significant difference in the
frequency of peri-implantitis related to the type of implant, except for a higher frequency of peri-implantitis
around implants with rough surfaces than around those
with smooth surfaces. There is limited and conflicting
evidence to suggest that implant surface and design
may be a risk indicator for peri-implantitis. 26 In this
study, differences in implant surface were not taken into
account, although it may be an important area for future
research.
In this study, the bacterial flora of peri-implantitis
sites was characterized using anaerobic culture techniques and 16S rDNA gene-based PCR for bacterial
identification, and compared with that of healthy implant sites. All bacterial strains were cultivated under obligate anaerobic conditions (redox potential:
400 mV). Culture methods are known to be useful
tools for studying the bacterial flora of infectious lesions by detecting which microorganisms inhabit a
given area,15 although the process is labor intensive
and time-consuming. However, sensitive and accurate
molecular techniques are necessary to characterize
the bacterial flora in peri-implantitis in order to determine their association with clinical symptoms and the
prognosis of treatment. In the past, a checkerboard
DNADNA hybridization technique was the most
usual method for identifying bacteria in the field of
clinical dentistry.27 Although this technique can be
used to detect gene-specific bacteria, it cannot randomly detect bacteria. Therefore, culture techniques
under obligate anaerobic conditions and 16S rDNA
gene-based PCR are important tools for detecting

oral bacterial species in the human oral cavity, and for

identifying the predominant bacteria that may be important pathogens for infection.
As a result, the present study has demonstrated significant differences between the bacterial flora in the
two groups, and has revealed characteristics of the
bacterial flora of peri-implantitis sites (Figs 5 and 6). In
addition, significantly more CFUs were observed in periimplantitis sites than in healthy implant sites (Table 2).
When the flora of the two groups were compared at
the genus level, the present study showed that the predominant genera in the peri-implantitis sites were Streptoccocus and Eubacterium, while genera Streptoccocus,
Veillonella, and Actinomyces were predominant in the
healthy implant sites (Figs 1 and 2).
In the literature there has been much speculation
surrounding the relationship between implant failures
and peri-implantitis, the flora of periodontitis, and the
flora of peri-implantitis. Failing oral implants are generally found to harbor bacterial flora traditionally associated with periodontitis such as P gingivalis, T forsythia,
T denticola, and F nucleatum. Previous studies have
found that the bacterial flora of peri-implantitis and
periodontal infections was similar,1,10 and one study
demonstrated that in several cases of peri-implantitis,
culture techniques identified species that included not
only pathogens such as P gingivalis, P intermedius, and
A actinomycetemcomitans, but also Escherichia coli and
S aureus.28
Although Streptococcus was one of the predominant genera detected in samples from both groups, the
species of the genus at the two sites were completely
different. S constellatus and S sobrinus were frequently
isolated from the peri-implantitis sites, but were not
detected at the healthy implant sites (Figs 5 and 6). In
contrast, S parasanguinis, S mutans, and S oralis were
detected only at healthy implant sites (Figs 5 and 6).
A long-term study on colonization of the peri-implant
area showed a decrease in the proportion of facultative anaerobic cocci such as Streptococcus species and
an increase in the percentage of obligate anaerobic
rods, eg, Fusobacterium spp and Prevotella spp.29 These
results may indicate the possibility that some species
of genus Streptococcus have ecological significance in
the flora of peri-implantitis sites.
Veillonella spp were found to be the predominant bacteria in healthy implant sites in our study. This finding is
consistent with the observations of the previous study. 7
OGNRs were detected in both groups. Long-term
clinical studies have suggested that the bacterial flora
of peri-implantitis sites was characterized by a high
proportion of OGNRs belonging to the red and orange
complex species.17,27,30,31 In this study, microbiological
culture of peri-implantitis sites has detected bacterial
species including P gingivalis, P intermedia, F nucleatum,
The International Journal of Oral & Maxillofacial Implants 1527

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

Parvimonas micra, and other major pathogenic bacteria characteristic of chronic periodontitis. Other studies
have reported that implants with peri-implantitis harbor A actinomycetemcomitans.32,33 However, these findings were not confirmed at peri-implantitis sites in this
study. Our results showed the presence of P gingivalis,
but red complex bacteria (T forsythia, T denticola),
A actinomycetemcomitans, S aureus, and Candida spp
were not detected. These bacteria may be cultivable but
undetectable for a mean CFU threshold level of 1104
per mL; however, these bacteria must certainly be present at lower levels than other red, orange, green, purple,
and yellow complexes.
On the other hand, gram-positive bacilli accounted
for 35% of the total proportional distribution at periimplantitis sites and 27% at healthy implant sites. In
addition, AAGPRs were detected in the peri-implantitis group (18%) but were almost undetectable in
the healthy implant group (3%), confirming that the
constitution of the bacterial flora differed markedly
between the two groups (Figs 3 and 4). A previous
study reported that many AAGPRs are significantly associated with periodontitis, periapical infections, and
oral abscesses, but are rarely found under healthy oral
conditions.13 In the present study, an overwhelming
number of AAGPRs (colored green in Fig 3) and OGNRs
(colored green in Fig 4) were detected in peri-implantitis
sites. These bacteria were obligate anaerobes and were
detected only in limited numbers at healthy implant
sites. The proportion of AAGPRs was almost the same
as the proportion of OGNRs at peri-implantitis sites.
However, AAGPRs were outnumbered more than two
to one at healthy implant sites. The ratio of AAGPRs
to OGNRs and total isolates increased significantly at
peri-implantitis sites. Uematsu and Hoshino reported
that AAGPRs, such as Eubacterium, Mogibacterium,
Slackia, and others, often predominated in periodontitis sites.14 Thus, it appears that AAGPR growth and biofilm formation may be influenced by periodontopathic
bacteria.
E nodatum, E saphenum, E minutum, and Filifactor
alocis, which are species of AAGPRs and were detected in
this study, have been known to produce the short-chain
fatty acid butyrate as one of the end products.14,15,18,34,35
OGNRs such as Fusobacterium spp and Prevotella spp
are also known to produce butyrate.3638 Butyrate has
been reported to inhibit cell growth of human gingival
fibroblasts39 and human endothelial cell proliferation in
vitro,40,41 and to induce apoptosis in T and B cells.37,42,43
Therefore, butyrate may have an important role in the
local tissue destructive process in the pathogenesis of
peri-implantitis. The presence of AAGPRs in this study
suggests that AAGPRs may possibly act in periodontitis
to cause tissue destruction at the base of the advanced
progressive peri-implantitis sulcus. This suggests that

bacterial flora at peri-implantitis sites reveals that

AAGPRs are also predominant in addition to specific
bacteria linked with peri-implantitis.
Previously, it was reported that uncultivated bacterial species were identified in peri-implantitis sites
by 16S rRNA gene clone library analysis.44 Since these
species might not be detected in the present study
with culture techniques, further studies are necessary
for considering the role of uncultivated or viable nonculturable bacterial species.

Conclusions
The present study indicates that an obligate anaerobic
environment exists at the base of the peri-implantitis
sulcus, and is suited for growth of AAGPRs and OGNRs.
Although further work is necessary to elucidate the
bacterial flora in peri-implantitis, the fact that the sulcus around oral implants with peri-implantitis showed
high levels of AAGPRs and OGNRs suggests that conventional periodontopathic bacteria are not the only
periodontal pathogens active in peri-implantitis, and
that AAGPRs may possibly play an important role.

Acknowledgments
The Health Sciences University of Hokkaido and the Health Sciences University of Hokkaido Hospital supported this work. The
authors reported no conflicts of interest related to this study.

References
1. Pye AD, Lockhart DE, Dawson MP, Murray CA, Smith AJ. A review of
dental implants and infection. J Hosp Infect 2009;72:104110.
2. Albrektsson T, Isidor E. Consensus report of session IV. In: Lang NP,
Karring T (eds). Proceedings of the First European Workshop on
Periodontology. London: Quintessence, 1994:365369.
3. Lindhe J, Meyle J. Peri-implant diseases: Consensus report of the
Sixth European Workshop on Periodontology. J Clin Periodontol
2008;35:282285.
4. Brgger U, Hugel-Pisoni C, Brgin WB, Buser D, Lang NP. Correlations
between radiographic, clinical, and mobility parameters after loading of oral implants with fixed partial dentures. A 2-year longitudinal
study. Clin Oral Implants Res 1996;7:230239.
5. Marinello CP, Berglundh T, Ericsson I, Klinge B, Glantz PO, Lindhe J.
Resolution of ligature-induced peri-implantitis lesions in the dog.
J Clin Periodontol 1995;22:475479.
6. Leonhardt , Renvert S, Dahln G. Microbial findings at failing
implants. Clin Oral Implants Res 1999;10:339345.
7. Frst MM, Salvi GE, Lang NP, Persson GR. Bacterial colonization
immediately after installation on oral titanium implants. Clin Oral
Implants Res 2007;18:501508.
8. Mombelli A, van Oosten MAC, Schrch E Jr, Lang NP. The microbiota
associated with successful or failing osseointegrated titanium
implants. Oral Microbiol Immunol 1987;2:145151.
9. Slots J, Rams TE. New views on periodontal microbiota in special
patient categories. J Clin Periodontol 1991;18:411420.
10. Shibli JA, Melo L, Ferrari DS, Figueiredo LC, Faveri M, Feres M. Composition of supra and subgingival biofilm of subjects with healthy
and diseased implants. Clin Oral Implants Res 2008;19:975982.

1528 Volume 28, Number 6, 2013

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Tamura et al

11. Leonhardt , Grndahl K, Bergstrm C, Lekholm U. Long-term

follow-up of osseointegrated titanium implants using clinical, radiographic, and microbiological parameters. Clin Oral Implants Res
2002;13:127132.
12. Nakazawa F, Hoshino E. Genetic relationships among Eubacterium
species. Int J Syst Bacteriol 1994;44:787790.
13. Wade WG, The role of Eubacterium species in periodontal disease
and other oral infections. Microb Ecol Health 1997;9:367370.
14. Uematsu H, Hoshino E. Predominant obligate anaerobes in human
periodontal pockets. J Periodont Res 1992;27:1519.
15. Poco SE, Nakazawa F, Sato M, Hoshino E. Eubacterium minutum sp.
nov., isolated from human periodontal pockets. Int J Syst Bacteriol
1996;46:3134.
16. Nakazawa F, Hoshino E. Immunological specificity of oral Eubacterium
species. J Gen Microbiol 1993;139:26352640.
17. Socransky SS, Haffajee AD. Dental biofilms: Difficult therapeutic
targets. Periodontol 2000 2002;28:1255.
18. Cheeseman SL, Hiom SJ, Weightman AJ, Wade WG. Phylogeny of oral
asaccharolytic Eubacterium species determined by 16S ribosomal
DNA sequence comparison and proposal of Eubacterium infirmum sp.
nov. and Eubacterium tardum sp. nov. Int J Syst Bacteriol 1996;46:
957959.
19. Lang NP, Nyman S, Senn C, Joss A. Bleeding on probing as it
relates to probing pressure and gingival health. J Clin Periodontol
1991;18:257261.
20. Lisa JA, Mayfield H. Peri-implant diseases: Diagnosis and risk indicators. J Clin Periodontol 2008;35:292304.
21. Mombelli A, Graf H. Depth-force pattern in periodontal probing.
J Clin Periodontol 1986;13:126130.
22. Huys G, Vanhoutte T, Joossens M, et al. Coamplification of eukaryotic
DNA with 16S rRNA gene-based PCR primers: Possible consequences for population fingerprinting of complex microbial communities.
Curr Microbiol 2008;56:553557.
23. Washio J, Sato T, Koseki T, Takahashi N. Hydrogen sulfide-producing
bacteria in tongue biofilm and their relationship with oral malodor.
J Med Microbiol 2005;54:889895.
24. Fox GE, Wisotzkey JD, Jurshuk P Jr. How close is close: 16S rRNA sequence
identity may not be sufficient to guarantee species identity. Int J Syst
Bacteriol 1992;42:166170.
25. Stackebrant E, Goebel BM. Taxonomic note: A place for DNA-DNA
reassociation and 16S rRNA sequence analysis in the present species
definition in bacteriology. Int J Syst Bacteriol 1994;44:846849.
26. Heitz-Mayfield LJA. Peri-implant diseases, diagnosis and risk indicators. J Clin Periodontol 2008;35(suppl 8):292304.
27. Socransky SS, Haffajee AD, Cugini MA, Smith C, Kent RL Jr. Microbial
complexes in subgingival plaque. J Clin Periodontol 1998;25:134144.
28. Leonhardt , Dahlen G, Renvert S. Five-year clinical, microbiological,
and radiological outcome following treatment of peri-implantitis in
man. J Periodontol 2003;74:14151422.
29. Mombelli A, Mericske-Stern R. Microbiological features of stable
osseointegrated implants used as abutments for overdentures. Clin
Oral Implants Res 1990;1:17.

30. Bower RC, Radny NR, Wall CD, Henry PJ. Clinical and microscopic findings
in edentulous patients 3 years after incorporation of osseointegrated
implant-supported bridgework. J Clin Periodontol 1989;16:580587.
31. Botero JE, Gonzalez AM, Mercado RA, Olave G, Contreras A. Subgingival microbiota in peri-implant mucosa lesions and adjacent teeth
in partially edentulous patients. J Periodontol 2005;76:14901495.
32. Van Winkelhoff AJ, Wolf JW. Actinobacillus actinomycetemcomitans
associated peri-implantitis in an edentulous patient. A case report.
J Clin Periodontol 2000;27:531535.
33. Hultin M, Gustafsson A, Hallstrom H, Johansson LA, Ekfeldt A, Klinge
B. Microbiological findings and host response in patients with periimplantitis. Clin Oral Implants Res 2002;13:349358.
34. Cato EP, Moore LVH, Moore WEC. Fusobacterium alocis sp. nov. and
Fusobacterium sulci sp. nov. from the human gingival sulcus. Int J
Syst Bacteriol 1985;35:475477.
35. Jalava J, Eerola E. Phylogenetic analysis of Fusobacterium alocis and
Fusobacterium sulci baced on 16S rRNA gene sequences: Proposal of
Filifactor alocis (Cato, Moore and Moore) comb. nov. and Eubacterium sulci (Cato, Moore and Moore) comb. nov. Int J Syst Bacteriol
1999;49:13751379.
36. Bartold PM, Cully NJ, Zilm PS, Rogers AH. Identification of components in Fusobacterim nucleatum chemostat-culture supernatants
that are potent inhibitors of human gingival fibroblast proliferation.
J Periodont Res 1991;26:314322.
37. Kurita OT, Ochiai K, Fukushima K. Volatile fatty acid, metabolic by
product of periodontopathic bacteria, induces apoptosis in WEHI
231 and RAJI B lymphoma cells and splenic B cells. Infect Immun
1998;66:25872594.
38. Takahashi N, Sato T, Yamada T. Metabolic pathways for cytotoxic end
product formation from glutamate and aspartate-containing peptides by Porphyromonas gingivalis. J Bacteriol 2000;182:47044710.
39. Jeng JH, Chan CP, Ho YS, Lan WH, Hsieh CC, Chang MC. Effects of
butyrate and propionate on the adhesion, growth, cell cycle kinetics, and protein synthesis of cultured human gingival fibroblasts.
J Periodontol 1999;70:14351442.
40. Tse CS, Williams DM. Inhibition of human endothelial cell proliferation in vitro in response to n-butyrate and propionate. J Periodont
Res 1992;27:506510.
41. Pllnen MT, Overman DO, Salonen JI. Bacterial metabolites sodium
butyrate and propionate inhibit epithelial cell growth in vitro.
J Periodont Res 1997;32:326334.
42. Kurita OT, Fukushima K, Ochiai K. Butyric acid-induced apoptosis of
murine thymocytes, splenic T cells, and human Jurkat T cells. Infect
Immun 1997;65:3541.
43. Kurita OT, Fukushima K, Ochiai K. Lipopolysaccharide stimulates
butyric acid-induced apoptosis in human peripheral blood mononuclear cells. Infect Immun 1999;67:2229.
44. Koyanagi T, Sakamoto M, Takeuchi Y, Ohkuma M, Izumi Y. Analysis
of microbiota associated with peri-implantitis using 16S rRNA gene
clone library. J Oral Microbiol 2010;2:51045110.

The International Journal of Oral & Maxillofacial Implants 1529

2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY.
NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.