Detection of five potentially periodontal pathogenic bacteria in peri-implant
disease: A comparison of PCR and real-time PCR

Gerhard Schmalz, Sandra Tsigaras, Sven Rinke, Tanja Kottmann, Rainer

Haak, Dirk Ziebolz

PII: S0732-8893(16)30079-7
DOI: doi: 10.1016/j.diagmicrobio.2016.04.003
Reference: DMB 14058

To appear in: Diagnostic Microbiology and Infectious Disease

Received date: 10 November 2015

Revised date: 7 March 2016
Accepted date: 5 April 2016

Please cite this article as: Schmalz Gerhard, Tsigaras Sandra, Rinke Sven, Kottmann
Tanja, Haak Rainer, Ziebolz Dirk, Detection of ve potentially periodontal pathogenic
bacteria in peri-implant disease: A comparison of PCR and real-time PCR, Diagnostic
Microbiology and Infectious Disease (2016), doi: 10.1016/j.diagmicrobio.2016.04.003

This is a PDF le of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its nal form. Please note that during the production process
errors may be discovered which could aect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Detection of five potentially periodontal pathogenic bacteria in peri-implant disease: A

comparison of PCR and real-time PCR

Gerhard Schmalz1, Sandra Tsigaras2, Sven Rinke3,4, Tanja Kottmann5, Rainer Haak1, Dirk
Ziebolz1

P T
1) Dept. of Cariology, Endodontology and Periodontology, University of Leipzig, Germany

RI
2) Dept. of Preventive Dentistry, Periodontology and Cariology, University Medical Centre
Goettingen, Germany

SC
3) Dental Practice Hanau & Alzenau, Germany
4) Dept. of Prosthodontics, University Medical Centre Goettingen, Germany

NU
5) Clinical Research Organisation, Hamm, Germany
MA
Short running title: Microbiological diagnosis of peri-implant disease

Word count abstract: 150

ED

Word count text: 3111

PT

Corresponding author:
PD Dr. Dirk Ziebolz, M.Sc.
University Medical Center Leipzig
CE

Dept. of Cariology, Endodontology and Periodontology

Liebigstr. 10-14
D 04103 Leipzig
Germany
AC

Tel.: +49 341 97 21211

Fax: +49 341 97 21219
mail: dirk.ziebolz@medizin.uni-leipzig.de

0
 ACCEPTED MANUSCRIPT

Abstract

The aim of this study was to compare the microbial analysis methods of polymerase chain
reaction (PCR) and real-time PCR (RT-PCR) in terms of detection of five selected potentially
periodontal pathogenic bacteria in peri-implant disease. Therefore 45 samples of healthy,

T
mucositis and peri-implantitis (n=15 each) were assessed according to presence of the

P
following bacteria using PCR (DNA-strip technology) and RT-PCR (fluorescent dye SYBR

RI
green-system): Aggregatibacter actinomycetemcomitans (Aa), Porphyromonas gingivalis
(Pg), Treponema denticola (Td), Tanerella forsythia (Tf), and Fusobacterium nucleatum (Fn).

SC
There were no significant correlations between the bacterial and disease patterns, so the
benefit of using microbiological tests for the diagnosis of peri-implant diseases is

NU
questionable. Correlations between the methods were highest for Tf (Kendall's Tau: 0.65,
Spearman: 0.78), Fn (0.49, 0.61) and Td (0.49, 0.59). For Aa (0.38, 0.42) and Pg (0.04,
MA
0.04), lower correlation values were detected. Accordingly, conventional semi-quantitative
PCR seems to be sufficient for analyzing potentially periodontal pathogenic bacterial species.
ED

Key words: peri-implant mucositis, peri-implantitis, microbiological findings, PCR, real-time

PT

PCR
CE
AC

1
 ACCEPTED MANUSCRIPT

1. Introduction
In recent years, dental implants for the replacement of missing teeth have exhibited
remarkable long-term stability. This assertion is confirmed by current studies that have
demonstrated survival and success rates greater than 90% for up to 10 years (Buser et al.,
2012, Filippi et al., 2013, Sanz et al., 2015). However, peri-implant complications are still

T
observed; thus, peri-implant mucositis (M) and peri-implantitis (P) remain important biological

P
complications (Mombelli et al., 2012).

RI
To avoid these complications and ensure high success rates, early diagnoses are of
paramount importance. Both M and P are inflammatory infectious diseases that are primarily

SC
caused by bacteria (Mombelli et al., 2012). Accordingly, microbiological analyses of P and M
could serve as an important method for early detection. For periodontitis, microbiological

NU
testing and monitoring appears to be helpful, especially in patients where the standard
therapy is not successful; however, strong evidence for benefit of microbiological testing for
MA
periodontal diseases is still missing (Listgarten and Loomer, 2003). Past studies have also
investigated peri-implant microbiota (Mombelli and Dcaillet, 2011) and reported greater
bacterial diversities than those observed in periodontitis biofilms, including staphylococci and
enteric bacteria (Belibasakis, 2014, Faveri et al., 2015). However, common periodontal
ED

pathogenic bacteria are repeatedly detected in peri-implant diseased conditions (Shibli et al.,
2008, Maximo et al., 2009, Persson and Renvert, 2014, Albertini et al., 2015). Furthermore,
PT

different results regarding the associations of potentially periodontal pathogenic bacteria with
M and P are available; thus, the actual opportunities and benefits remain unclear (Mombelli
CE

and Dcaillet, 2011, Charalampakis and Belibasakis, 2015). Therefore, further investigations
appear to be necessary to gain insight into the real microbiological diagnostic opportunities
for peri-implant diseases. The potential advantages could include early diagnosis and
AC

determination of risk adapted maintenance programs based on microbiological

findings,which justifies further investigation of this issue.
Different procedures for the microbiological diagnosis of potentially periodontal pathogenic
bacteria are available, including polymerase chain reaction (PCR), which has produced exact
results in the available studies of periodontitis and peri-implantitis (Haffajee et al., 2009,
Ertugrul et al., 2013, Albertini et al., 2015, Eick et al., 2015). With the introduction of
commercial tests based on PCR, the qualitative and semi quantitative detection of different
bacteria with low temporal expenditure has become possible. An advancement of this
process is real-time polymerase chain reaction (RT-PCR), which allows for the quantitative
detection of bacterial DNA using fluorescence signals and exhibits high specificity and also
successful results regarding the detection of potentially periodontal pathogenic bacteria
(Socransky et al., 2005, Boutaga et al., 2007). The available studies have also demonstrated
that the bacteria of peri-implant biofilms are detectable with both PCR and RT-PCR analyses

2
 ACCEPTED MANUSCRIPT

(Sato et al., 2011, Al-Radha et al., 2012, Galassi et al., 2012, Zhuang et al., 2014, Canullo et
al., 2015). However, it is questionable whether conventional PCR provides sufficiently exact
results and whether RT-PCR could benefit the microbiological diagnosis of periodontal and
peri-implant disease. A recent study compared these methods in terms of periodontitis and
found no significant differences, i.e., neither procedure was found to be superior (Eick et al.,

T
2011). Comparable data regarding M and P are already not available, yet.

P
Accordingly, the current study was performed to investigate whether PCR and RT-PCR are

RI
able to show differences in detection of five common potentially periodontal pathogenic
bacteria between healthy, M and P peri-implant conditions. A further aim was to compare

SC
PCR and RT-PCR analyses to determine whether one of these methods is substantially
preferable for the microbiological analyses of M and P. The hypothesis of this study was that

NU
conventional PCR provides sufficient results for the detection of selected potentially
periodontal pathogenic bacteria in peri-implant disease.
MA
2. Methods
2.1. Study design
This clinical-experimental examination was performed to investigate different microbiological
ED

diagnostic procedures (i.e., PCR and RT-PCR) for peri-implant disease. The biofilm samples
originated from the patient clientele of a clinical study from this working group (Rinke et al.,
PT

2011). The study protocol was reviewed and approved by the ethics committee of the
University Medical Center Goettingen, Germany (No. 3/1/09). The patients were informed
CE

about the study and provided written informed consent.

2.2. Patients
From a pool of 171 peri-implant biofilm samples from 89 patients, 45 were selected for
AC

analysis. The implants were classified as healthy or with mucositis or peri-implantitis

according the method detailed in a previous publication by this working group (Rinke et al.,
2011). Forty-five samples were randomly divided in the three groups (15 samples in each
group) as follows: 15 different samples from healthy implants from implant carriers without
further implants in the mouth, 15 different samples from implants with mucositis from implant
carriers without further diseased implants in the mouth, and 15 different samples from
implants with peri-implantitis with partially samples from additional implants from the same
patients. The number of samples investigated conformed to the number of peri-implantitis
samples, which were 15 in the patients clientele. Therefore it was not possible to include
more than 15 samples each group.
2.3. Sample collection and DNA isolation

3
 ACCEPTED MANUSCRIPT

Subgingival biofilm samples were collected using sterile paper points. After removing the
supragingival biofilm, the implant was drained using cotton rolls. Paper points were placed in
the mesial and distal pocket fundi for 10 s, pooled and placed in a transportation tube.
For the detection of potentially periodontal pathogenic bacteria, the isolation of DNA is
necessary. To achieve this goal, QIAamp DNA Mini Kits (Qiagen, Hilden, Germany) were

T
used according to the manufacturers instructions. Subsequently, the samples were stored at

P
-20C until further processing.

RI
2.4. Polymerase chain reaction (PCR)
PCR was performed in the laboratory of the Department of Preventive Dentistry,

SC
Periodontology and Cariology of the University Medical Centre Goettingen, Germany. The
IDent and Micro-IDent plus (Hain Lifescience, Nehren, Germany) commercial test systems

NU
were used for the qualitative and semiquantitative detections of the potentially periodontal
pathogenic bacteria according to manufacturers protocol. In brief, amplification was
MA
performed using a 35-l mixture of primers and dNTPs (Hain Lifescience, Nehren, Germany),
10.5 l Mastermix (Qiagen, Hilden, Germany) and 5 l of the DNA sample or 5 l of water as
a negative control. The cycles were executed in a thermo cycler (Biometra, Goettingen,
Germany). Initially, one cycle was conducted for 5 min at 95C, subsequently 10 cycles of 30
ED

s at 95C and 2 min at 58C followed by 20 cycles of 25 s at 95C, 40 s at 53C and 40 s at

70C and a final cycle of 8 min at 70C were performed. Hybridization was executed
PT

according to the Micro-IDent plus (Hain Lifescience, Nehren, Germany) protocol. In brief,
after denaturation and biotinylation, the samples were incubated with hybridization buffer and
CE

probes containing membrane strips at 45C in a TwinCubator (Hain Lifescience, Nehren,

Germany). The membrane strips also contained two control lines (i.e., a conjugate control
and an amplification control). Next, highly specific washing steps were executed to remove
AC

the nonspecifically bound DNA. Each membrane strip was incubated with a streptavidin-
alkaline phosphatase complex containing conjugate and intensively washed. The color
reactions induced by the reaction between alkaline phosphatase and the substrate on the
membrane strips with bound amplification product was registered. Accordingly, the
qualitative and semi-quantitative detections of Aggregatibacter actinomycetemcomitans (Aa),
Porphyromonas gingivalis (Pg), Treponema denticola (Td), Tannerella forsythia (Tf) and
Fusobacterium nucleatum (Fn) were performed based on the colored bands (relative
concentrations <10 to >107).
2.5. Real-time PCR
Before the RT-PCR was executed, preliminary tests were conducted to detect reliable
primers for Aa, Pg, Td, Tf and Fn. Following literature research, whether the primers were
able to amplify the DNA of the patient samples was investigated, and water samples
simultaneously served as negative controls. The results were validated using the

4
 ACCEPTED MANUSCRIPT

fluorescence signals from real-time PCR and agarose gel electrophoresis. The primers used
in this study are listed in table 1.
For exact results, different controls were used. One reference pool containing a mixture of all
45 samples was used as a reference for relative quantification. Moreover, a mixture
containing 19 DNA samples with known positive findings for the investigated bacteria from

T
parallel studies was used as a positive reference. Additionally, all samples were analyzed in

P
triplicate. Real-time PCR was performed in a thermocycler (Bio-Rad, Munich, Germany)

RI
using 5 l sample and 20 l Mastermix containing 12.5 l SYBRgreen Supermix, 6.5 l
Ampuwa (Fresenius, Bad Homburg, Germany), 0.5 l sense primer and 0.5 l antisense

SC
primer. Following centrifugation for 3 min at 1500 x g, the samples were amplified using the
following conditions: initial denaturation for 5 min at 95C; 45 cycles of 15 s at 95C, 20 s at

NU
60C (Aa: 58C) and 20 s at 72C; a cycle of 1 min at 95C and 1 min at 55C; and 75 cycles
of 10 s at 55C. The evaluation of the RT-PCR was executed as a relative quantification in
MA
which a pool of 45 probes was set as 100% with a value of 2. Accordingly, the values for the
concentrations of bacteria ranged between 0 and 2.
Additionally, agarose gel electrophoresis was performed to validate the RT-PCR results. A
3% agarose gel containing 2.4 g agarose (Roth, Karlsruhe, Germany), 80 ml Tris-acetate-
ED

EDTA buffer (TAE buffer, 1x, 1:50 with Aqua destillata) was used. Five microliters of each
RT-PCR sample was mixed with 1 l loading buffer (Roth, Karlsruhe, Germany) containing
PT

glycerol and bromophenol blue for color detection. One sample containing a pool of 45
samples, single positive and negative control samples, and randomly chosen RT-PCR
CE

samples were investigated. Standardized 5 l 20 bp and 100 bp ladders (Roth, Karlsruhe,

Germany) were used as references. The gel electrophoresis was executed in an
electrophoresis chamber (Polymehr, Paderborn, Germany) at 85 V for 1 h. Subsequently, the
AC

samples were dyed with ethidium bromide (Roth, Karlsruhe, Germany) and visualized using
a BioDoc Analyzer (Biometra, Goettingen, Germany).
2.6. Statistical analysis
The Statistica software (StatSoft, Hamburg, Germany, version 10, 2010) was used for the
statistical analyses. Fishers exact tests were performed (p<0.05) to detect significant results
between the bacteria and disease based on PCR. The significant results between disease
and bacteria based on RT-PCR were determined with ANOVA (p<0.05). Spearman and
Kendall Tau tests were used to detect significant correlations between the PCR and RT-PCR
results (p<0.05). Furthermore, sensitivity and specificity was determined for both PCR and
RT-PCR results.

3. Results
3.1. PCR analysis

5
 ACCEPTED MANUSCRIPT

All investigated bacteria (i.e., Aa, Pg, Td, Tf and Fn) were detected in the healthy, M and P
peri-implant conditions. Detailed concentration and prevalence values are provided in table
2. The PCR analysis did not reveal any significant results for Aa (p=0.56), Pg (p=0.2), Td
(p=0.31), Tf (p=0.71) and Fn (p=0.41) regarding healthy (H), mucositis (M) or peri-implantitis
(P) conditions (figure 1a).

T
3.2. Real-time PCR analysis

P
The RT-PCR analysis detected all investigated potentially periodontal pathogenic bacteria

RI
(i.e., Aa, Pg, Td, Tf and Fn) equally in the healthy, M and P peri-implant conditions.
Accordingly, no significant results for Aa (p=0.32), Pg (p=0.16), Td (p=0.92), Tf (p=0.97) or

SC
Fn (p=0.87) with regard to peri-implant disease were identified (figure 1b). Detailed
concentration and prevalence values are given in table 2.

NU
3.3. Comparison of PCR and RT-PCR
All 45 samples were investigated with both of the compared methods. Both the PCR and RT-
MA
PCR analyses detected all investigated potentially periodontal pathogenic bacteria in the
healthy, M and P peri-implant conditions (tables 2 and 3). Regarding prevalence, what
means the number of positive findings with regard to the number of investigated samples,
PCR and RT-PCR produced similar values for Aa (20% and 22%), Tf (55% and 53%) and Fn
ED

(89% and 100%), whereas these methods exhibited greater differences for Pg (67% and
38%) and Td (33% and 55%; table 4). Exact matches the concentrations detected by each
PT

method were observed for Aa (84%), Pg (33%), Td (51%), Tf (55%) and Fn (29%; table 4).
The correlations between the methods were highest for Tf (Kendalls Tau: 0.65, Spearman:
CE

0.78), Fn (Kendalls Tau: 0.49, Spearman: 0.61) and Td (Kendalls Tau: 0.49, Spearman:
0.59). Lower correlations were detected for Aa (Kendalls Tau: 0.38, Spearman: 0.42) and Pg
(Kendalls Tau: 0.04, Spearman: 0.04; table 4). Sensitivity and specificity of PCR and RT-
AC

PCR for discrimination between healthy and mucositis as well as healthy and peri-implantitis
for the five investigated bacteria is shown in table 5.

4. Discussion
Summary of the main results: All investigated potentially periodontal pathogenic bacteria
(i.e., Aa, Pg, Td, Tf and Fn) were detected in the healthy, M and P peri-implant conditions by
both methods. Significant results for bacteria regarding peri-implant disease were not
detected. The PCR- RT-PCR comparison revealed that both methods produced similar
results.
Comparison with the available literature and interpretation of the results: In the current study,
all detected potentially periodontal pathogenic bacteria were observed with similar
prevalences in the healthy, M and P peri-implant conditions. In accordance with these
results, several results have been reported that have indicated a lack of reliable differences
between healthy and diseased implants in terms of their bacterial populations (Dabdoub et

6
 ACCEPTED MANUSCRIPT

al., 2013, Koyanagi et al., 2013, Zhuang et al., 2014, Canullo et al., 2015). Interestingly, the
majority of these studies also performed RT-PCR for detection (Koyanagi et al., 2013,
Zhuang et al., 2014, Canullo et al., 2015). In contrast however, different studies are available
that have demonstrated a higher prevalence of potentially periodontal pathogenic bacteria in
peri-implantitis. A portion of these studies used DNA-DNA checkerboard hybridization as the

T
detection method (Shibli et al., 2008, Maximo et al., 2009, Persson and Renvert, 2014).

P
Additional studies have performed PCR for the detection of potentially periodontal

RI
pathogenic bacteria in peri-implant disease (Albertini et al., 2015, Eick et al., 2015). In this
context, Pg, Tf and Td have been detected, whereas Aa was not found in P (Albertini et al.,

SC
2015). Moreover, Eick et al. (2015) reported associations of Td with PPD and inflammation
(BOP and GI) on implants (Eick et al., 2015). This was not confirmed in the PCR analysis of

NU
the current study. In addition to potentially periodontal pathogenic bacteria, many studies
have detected additional microbiota that are potentially associated with P among which
MA
Staphylococcus ssp. might be the most common (Shibli et al., 2008 Maximo et al., 2009,
Persson and Renvert, 2014, Eick et al., 2015). However, the potential roles of these
microbes remain unclear (Belibasakis, 2014). As presented in the literature, the findings
ED

related to peri-implant biofilms are heterogeneous, and the real potential is not completely
clear (Faveri et al., 2015, Charalampakis and Belibasakis, 2015). For this reason, the current
study focused on common potentially periodontal pathogenic bacteria. Therefore, in relation
PT

to Socransky et al, Pg, Td and Tf were chosen as major pathogenes for periodontal diseases
(Socransky et al., 1998). Additionally, Aa which might be associated to aggressive
CE

periodontal diseases (Knnen and Mller, 2014) and Fn as a 'bridge-organism' with an

important role in formation of potentially periodontal pathogenic biofilm (Sharma et al., 2005)
AC

were investigated. All of these bacteria were already detected in peri-implant diseases,
however results were inconsistent (Shibli et al., 2008, Maximo et al., 2009, Persson and
Renvert, 2014, Albertini et al., 2015).
However, in the interpretation of the results of the current study, the facts that only five
potentially periodontal pathogenic bacteria were investigated in only 45 samples must be
taken into account. In comparable studies, the sample sizes have ranged between 44 and
504 peri-implant samples (Shiblin et al., 2008, Eick et al., 2015). Nevertheless, detections of
all investigated potentially periodontal pathogenic bacteria in the healthy, P and M peri-
implant conditions casts doubt on the diagnostic benefit. Consequently, based on the current
literature and the results of the present study, there appears to be no clear advantage of
microbiological testing for peri-implant disease at the moment.
An additional aspect is the comparison of PCR and RT-PCR in the detection of potentially
periodontal pathogenic bacteria in M and P. Conventional PCR, particularly that using Micro-
IDent plus (Hain Lifescience, Nehren, Germany) has been repeatedly and successfully

7
 ACCEPTED MANUSCRIPT

performed in periodontal and peri-implant disease in the available studies and is therefore
thought to be a valid method (Haffajee et al., 2009, Ertugrul et al., 2013, Eick et al., 2015).
Moreover, RT-PCR allows for the effective detection of potentially periodontal pathogenic
bacteria as demonstrated in recent investigations (Sato et al., 2011, Galassi et al., 2012,
Zhuang et al., 2014, Canullo et al., 2015). The current study revealed a very low correlation

T
between PCR and RT-PCR for Pg (0.04). PCR revealed a greater prevalence of Pg,

P
particularly in the healthy peri-implant conditions. Furthermore, Pg (33%) and Fn (29%)

RI
exhibited much weaker matches than Aa (84%), Td (51%) and Tf (55%). Different
methodological reasons could potentially be responsible for these results. For example, the

SC
RT-PCR was performed entirely in a thermocycler, whereas the PCR was partly performed in
a shaking bath and was thus associated with a greater risk of contamination. Additionally, the

NU
PCR primers used in these tests have not been undisclosed by the manufacturer, whereas
the RT-PCR primers were designed based on literature research (table 1) and validated in a
MA
preliminary experiment. Consequently, these methodological differences might be
responsible for the differences between the methods.
To the authors knowledge, there is no available study that has compared these methods in
terms of peri-implant disease. Regarding periodontitis, one investigation demonstrated
ED

significant correlations between both methods for Aa (0.68), Pg (0.74), Td (0.62), Tf (0.69)
and Fn (0.59) (Eick et al., 2011). Eick et al. (2011) also concluded that conventional PCR is
PT

more suitable and useful for dental practice. These results are in accordance with the results
of the current study, particularly in terms of the similar observed correlations between PCR
CE

and RT-PCR for Tf (0.78), Fn (0.61) and Td (0.59). The correlation coefficients for Aa (0.42)
and Pg (0.04) observed in the current study were clearly lower than those reported by Eick et
al. (2011). These differences might have been caused by the use of different primers.
AC

Additional studies comparing PCR and RT-PCR are not available for either peri-implantitis or
periodontitis. However, several studies have investigated PCR or RT-PCR compared with
checkerboard hybridization or cultivation; these studies demonstrated that PCR is the
predominant method (Riggio et al., 1996, Eick and Pfister, 2002, Haffajee et al., 2009).
Furthermore, a meta-analysis compared the detections of Aa and Pg by RT-PCR and
cultivation and found that RT-PCR is more exact (Atieh, 2008). In contrast, Lau et al. (2004)
reported a good match between RT-PCR and cultivation (Lau et al., 2004).
Therefore, the method should be chosen according to the aim of the investigation. RT-PCR
ensures the fast quantification of potentially periodontal pathogenic bacteria, whereas
antibiogram cultivation is necessary for the detection of bacterial activity (Verner et al., 2006).
For the semi-quantitative detection of microbiota in dental practice, commercial test systems
based on PCR appear to be sufficient. Accordingly, our hypothesis that conventional PCR
provides sufficient results the diagnoses of peri-implant diseases is confirmed.

8
 ACCEPTED MANUSCRIPT

Nevertheless, the results depend on the technique and test system and should be interpreted
only with consideration for the clinical diagnosis (Untch and Schlagenhauf, 2015). Taking the
results for analysis of sensitivity and specificity into account, the benefit of both PCR and RT-
PCR appears questionable. Heterogeneous and low values indicate an insufficient
distinctness between healthy and diseased peri-implant conditions for both methods. For

T
periodontal diseases, Eick et al., 2011 found higher sensitivity and specificity. Therefore, the

P
benefit of investigation of potentially periodontal pathogenic bacteria in periodontal diseases

RI
appears to be higher compared to peri-implant diseases.
Strengths and limitations: To the authors knowledge, this is the first study to compare PCR

SC
and RT-PCR in the detection of peri-implant disease. Furthermore, the detailed analyses of
the potentially periodontal pathogenic bacteria in the healthy, M and P peri-implant conditions

NU
are a strength of this study. The study was limited by small sample size of 45. However,
smaller sample sizes have been used to detect potentially periodontal pathogenic bacteria;
MA
so e.g., De Leito et al. (2005) investigated 19 samples (De Leito et al., 2005). The
investigation of more than one sample of patients with P was also suboptimal; however, the
patient clientele consisted of only very few patients with P. Furthermore, methodological
limitations regarding diagnosis and sample collection must be mentioned. The clinically
ED

diagnosis of H and M sites could be a source of flaws; however the diagnosis was executed
with highest care. Similarly, the sample collection in H and M sites is difficult, what should be
PT

considered in the interpretation of results.

The use of different primers may also have negatively influenced the correlation results.
CE

Especially, the primer sequences of the commercial test systems IDent and Micro-IDent plus
(Hain Lifescience, Nehren, Germany) are not given by the manufacturer. This must be taken
into account in interpretation of the results. A solution for future studies could be the use of
AC

arbitrarily primed PCR (AP-PCR) (Ito, 2010). With this method, the identification of exact
bacterial genotypes the accordingly specific primer design might ensure better comparability.
An additional strength of this study is that the same sample was investigated with both
methods. The performance of the experiments in triplicate and the preliminary test also
increased the quality of the current study.

Conclusion
Within the limitations of the current study, the similar PCR and RT-PCR results suggest that
neither method is superior. Accordingly, PCR-based test systems are sufficient for dental
practice. The detection of similar levels of potentially periodontal pathogenic bacteria in the
healthy, M and P peri-implant conditions makes the advantages of microbiological testing for
peri-implant disease questionable. Based on microbiological findings of potentially

9
 ACCEPTED MANUSCRIPT

periodontal pathogenic bacteria it is impossible to diagnose peri-implant diseases. Clinical

diagnostic of M and P remain indispensable.

Acknowledgments
The authors thanks Mrs. M. Hoch for support in the laboratory microbiological analysis.

P T
Conflict of interest and source of funding statement

RI
The authors declare that they have no financial or other relationships that might lead to a
conflict of interest.

SC
NU
MA
ED
PT
CE
AC

10
 ACCEPTED MANUSCRIPT

References

Albertini, M., Lpez-Cerero, L., O'Sullivan, M.G., Chereguini, C.F., Ballesta, S., Ros, V.,
Herrero-Climent, M. & Bulln, P. (2015) Assessment of periodontal and opportunistic flora in
patients with peri-implantitis. Clin Oral Implants Res 26: 937-941.
Al-Radha, A.S.D., Pal, A., Pettemerides, A.P. & Jenkinson, H.F. (2012) Molecular analysis of

T
microbiota associated with peri-implant diseases. J Dent 40: 989-998.

P
Ashimoto, A., Chen, C., Bakker, I. & Slots, J. (1996) Polymerase chain reaction detection of
8 putative periodontal pathogens in subgingival plaque of gingivitis and advanced

RI
periodontitis lesions. Oral Microbiol Immunol 11: 266-273.
Atieh, M.A. (2008) Accuracy of real-time polymerase chain reaction versus anaerobic culture

SC
in detection of aggregatibacter actinomycetemcomitans and porphyromonas gingivalis: a
meta-analysis. J Periodontol 79: 16201629.
Belibasakis, G.N. (2014) Microbiological and immuno-pathological aspects of peri-implant

NU
diseases. Arch Oral Biol 59: 66-72.
Boutaga, K., Savelkoul, P.H.M., Winkel, E.G. & Van Winkelhoff, A.J. (2007) Comparison of
subgingival bacterial sampling with oral lavage for detection and quantification of periodontal
pathogens by real-time polymerase chain reaction. J Periodontol 78: 7986.
MA
Braga, R.R.R., Carvalho, M.A.R., Brua-Romero, O., Teixeira, R.E., Costa, J.E., Mendes,
E.N., Farias, L.M. & Magalhes, P.P. (2010) Quantification of five putative periodontal
pathogens in female patients with and without chronic periodontitis by real-time polymerase
chain reaction. Anaerobe 16: 234239.
ED

Buser, D., Janner, S.F., Wittneben, J.G., Brgger, U., Ramseier, C.A. & Salvi, G.E. (2012)
10-year survival and success rates of 511 titanium implants with a sandblasted and acid-
etched surface: a retrospective study in 303 partially edentulous patients. Clin Implant Dent
PT

Relat Res 14: 839-851.

Canullo, L., Pearrocha-Oltra, D., Covani, U., Botticelli, D., Serino, G. & Penarrocha, M.
(2015) Clinical and microbiological findings in patients with peri-implantitis: a cross-sectional
CE

study. Clin Oral Implants Res 26. doi: 10.1111/clr.12557. [Epub ahead of print]
Charalampakis, G. & Belibasakis, G.N. (2015) Microbiome of peri-implant infections: lessons
from conventional, molecular and metagenomic analyses. Virulence 6: 183-187.
AC

Dabdoub, S.M., Tsigarida, A.A. & Kumar, P.S. (2013) Patient-specific analysis of periodontal
and peri-implant microbiomes. J Dent Res 92: 168-175
De Boever, A.L. & De Boever, J.A. (2006) Early colonization of non-submerged dental
implants in patients with a history of advanced aggressive periodontitis. Clin Oral Implants
Res 17: 8-17.
De Leito, J.A.O., De Lorenzo, J.L., Avila-Campos, M.J. & Sendyk, W.R. (2005) Analysis of
the presence of pathogens which predict the risk of disease at peri-implant sites through
polymerase chain reaction (PCR). Braz Oral Res 19: 5257.
Eick, S. & Pfister, W. (2002) Comparison of microbial cultivation and a commercial PCR
based method for detection of periodontopathogenic species in subgingival plaque samples.
J Clin Periodontol 29: 638644.
Eick, S., Straube, A., Guentsch, A., Pfister, W. & Jentsch, H. (2011) Comparison of real-time
polymerase chain reaction and DNA-strip technology in microbiological evaluation of
periodontitis treatment. Diagn Microbiol Infect Dis 69: 1220.
Eick, S., Ramseier, C.A., Rothenberger, K., Brgger, U., Buser, D. & Salvi, G.E. (2015)
Microbiota at teeth and implants in partially edentulous patients. A 10-year retrospective
study. Clin Oral Implants Res doi: 10.1111/clr.12588. [Epub ahead of print]

11
 ACCEPTED MANUSCRIPT

Ertugrul, A.S., Arslan, U., Dursun, R. & Hakki, S.S. (2013) Periodontopathogen profile of
healthy and oral lichen planus patients with gingivitis or periodontitis. International J Oral Sci
5: 92-97.
Faveri, M., Figueiredo, L.C., Shibli, J.A., Prez-Chaparro, P.J. & Feres, M. (2015)
Microbiological diversity of peri-implantitis biofilms. Adv Exp Med Biol 830: 85-96.
Filippi, A., Higginbottom, F.L., Lambrecht, T., Levin, B.P., Meier, J.L., Rosen, P.S.,

T
Wallkamm, B., Will, C. & Roccuzzo, M. (2013) A prospective noninterventional study

P
todocument implant success and survival of the Straumann Bone Level SLActive dental
implant in daily dental practice. Quintessence Int 44: 499-512.

RI
Fouad, A.F., Barry, J., Caimano, M., Clawson, M., Zhu, Q., Carver, R., Hazlett, K. & Radolf,
J.D. (2002) PCR-Based Identification of Bacteria Associated with endodontic infections. J

SC
Clin Microbiol 40: 32233231.
Galassi, F., Kaman, W.E., Anssari Moin, D., Van Der Horst, J., Wismeijer, D., Crielaard, W.,
Laine, M.L., Veerman, E.C., Bikker, F.J. & Loos, B.G. (2012) Comparing culture, real-time

NU
PCR and fluorescence resonance energy transfer technology for detection of
Porphyromonas gingivalis in patients with or without peri-implant infections. J Periodontal
Res 47: 616-625.
Haffajee, A.D., Yaskell, T., Torresyap, G., Teles, R. & Socransky, S.S. (2009) Comparison
MA
between polymerase chain reaction-based and checkerboard DNA hybridization techniques
for microbial assessment of subgingival plaque samples. J Clin Periodontol 36: 642649.
Ito, H. (2010) Development of a cps-based multiplex PCR for typing of Actinobacillus
pleuropneumoniae serotypes 1, 2 and 5. J Vet Med Sci 72: 653655.
ED

Koyanagi, T., Sakamoto, M., Takeuchi, Y., Maruyama, N., Ohkuma, M. & Izumi, Y. (2013)
Comprehensive microbiological findings in peri-implantitis and periodontitis. J Clin
Periodontol 40: 218-226.
PT

Knnen E. & Mller H.P. (2014) Microbiology of aggressive periodontitis. Periodontol 2000
65: 46-78.
CE

Lau, L., Sanz, M., Herrera, D., Morillo, J.M., Martn, C. & Silva, A. (2004) Quantitative real-
time polymerase chain reaction versus culture: a comparison between two methods for the
detection and quantification of Actinobacillus actinomycetemcomitans, Porphyromonas
gingivalis and Tannerella forsythensis in subgingival plaque samp. J Clin Periodontol 31:
AC

10611069.
Listgarten MA, Loomer PM (2003). Microbial identification in the management of periodontal
diseases. A systematic review. Ann Periodontol 8: 182-192.
Mximo, M.B., de Mendona, A.C., Renata Santos, V., Figueiredo, L.C., Feres, M. & Duarte,
P.M. (2009) Short-term clinical and microbiological evaluations of peri-implant diseases
before and after mechanical anti-infective therapies. Clin Oral Implants Res 20: 99-108.
Mombelli, A. & Dcaillet, F. (2011) The characteristics of biofilms in peri-implant disease. J
Clin Periodontol 11: 203-213.
Mombelli, A., Mller, N. & Cionca, N. (2012) The epidemiology of peri-implantitis. Clin Oral
Implants Res 23: 6776
Persson, G.R. & Renvert, S. (2014) Cluster of bacteria associated with peri-implantitis. Clin
Implant Dent Relat Res 16: 783-793.
Riggio, M.P., Macfarlane, T.W., Mackenzie, D., Lennon, A., Smith, A.J. & Kinane, D. (1996)
Comparison of polymerase chain reaction and culture methods for detection of Actinobacillus
actinomycetemcomitans and Porphyromonas gingivalis in subgingival plaque samples. J
Periodont Res 31: 496501.

12
 ACCEPTED MANUSCRIPT

Sanz, M., Ivanoff, C.J., Weingart, D., Wiltfang, J., Gahlert, M., Cordaro, L., Ganeles, J.,
Bragger, U., Jackowski, J., Martin, W.C., Jung, R.E., Chen, S. & Hammerle, C. (2015)
Clinical and radiologic outcomes after submerged and transmucosal implant placement with
two-piece implants in the anterior maxilla and mandible: 3-year results of a randomized
controlled clinical trial. Clin Implant Dent Relat Res 17: 234-246.
Sato, J., Gomi, K., Makino, T., Kawasaki, F., Yashima, A., Ozawa, T., Maeda, N. & Arai, T.

T
(2011) The evaluation of bacterial flora in progress of peri-implant disease. Aust Dent J 56:
201206.

P
Sharma A., Inagaki S., Sigurdson W., Kuramitsu H.K. (2005) Synergy between Tannerella

RI
forsythia and Fusobacterium nucleatum in biofilm formation. Oral Microbiol Immunol 20: 39-
42.

SC
Shibli, J.A., Melo, L., Ferrari, D.S., Figueiredo, L.C., Faveri, M. & Feres, M. (2008)
Composition of supra- and subgingival biofilm of subjects with healthy and diseased
implants. Clin Oral Implants Res 19: 975-982.

NU
Socransky S.S., Haffajee A.D., Cugini M.A., Smith C., Kent R.L. (1998) Microbial complexes
in subgingival plaque. J Clin Periodontol 25: 134144
Socransky, S.S., Sigmund, S. & Haffajee, A.D. (2005) Periodontal microbial ecology.
Periodontol 2000 38: 135187.
MA
Untch, M. & Schlagenhauf U. (2015) Inter- and intra-test agreement of three commercially
available molecular diagnostic tests for the identification of periodontal pathogens. Clin Oral
Invest 15. [Epub ahead of print]
ED

Verner, C., Lemaitre, P., Daniel, A., Giumelli, B., Lakhssassi, N. & Sixou, M. (2006)
Carpegen real-time polymerase chain reaction vs. anaerobic culture for periodontal pathogen
identification. Oral Microbiol Immunol 21: 341346.
PT

Zhuang, L.F., Watt, R.M., Mattheos, N., Si, M.S., Lai, H.C. & Lang, N.P. (2014) Periodontal
and peri-implant microbiota in patients with healthy and inflamed periodontal and peri-implant
tissues. Clin Oral Implants Res doi: 10.1111/clr.12508. [Epub ahead of print]
CE
AC

13
 ACCEPTED MANUSCRIPT

Legends

Table

Table 1: Primer sequences for real-time PCR, which were detected in a preliminary test and

T
used for the analysis.

P
primer design fragment Annealing

RI
length temperature
(bp) (C)

SC
forward 5-GTA CGG CGA AGG TAT TTC CA-3 197 58
Aa
backward 5-CCA ACA ATT CAT CAC GCA AG-3

Pg forward 5-GCG ACG CTA TAT GCA AGA CA-3 191 60

NU
backward 5- GCT GAT GGT GGC ATT ACC TT-3

Td forward 5-TAA TAC CGA ATG TGC TCA TTT ACA T-3 316 60
MA
backward 5-TCA AAG AAG CAT TCC CTC TTC TTC TTA-3
Tf MutaGel Periodontitis Kit (Immundiagnostik Bensheim, 120 60
Germany)
ED

Fn forward 5-CGC AGA AGG TGA AAG TCC TGT AT-3 23 60

backward 5-TGG TCC TCA CTG ATT CAC ACA GA-3
Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td: Treponema denticola, Tf: Tanerella forsythia,
PT

Fn: Fusobacterium nucleatum

CE
AC

14
 ACCEPTED MANUSCRIPT

Table 2: PCR analysis results for the investigated potentially periodontal pathogenic
bacteria. Absolute numbers of samples with the corresponding detection limit is given first
and in brackets values are given as % of the 15 investigated samples. The prevalence shows
the general number of positive results in relation to the 15 total samples.

T
Under detection limit Over detection limit
Concentrations prevalence

P
<10 10 <104 <105 >106

RI
H 13 (86) 1 (7) 1 (7) 0 (0) 0 (0) 2 (13)

Aa M 11 (73) 1 (7) 0 (0) 0 (0) 3 (20) 4 (27)

SC
P 12 (80) 2 (13) 0 (0) 0 (0) 1 (7) 3 (20)

Concentrations <104 104 <105 <106 >107

Pg
H

M
3 (20)

7 (47)
NU
5 (33)

2 (13)
2 (13)

0 (0)
2 (13) 3 (20)

2 (13) 4 (27)
12 (80)

8 (53)
MA
P 5 (33) 0 (0) 1 (7) 5 (33) 4 (2) 10 (67)

H 11 (73) 3 (20) 1 (7) 0 (0) 0 (0) 4 (27)

ED

Td M 11 (73) 1 (7) 0 (0) 3 (20) 0 (0) 4 (27)

P 8 (53) 4 (27) 1 (7) 1 (7) 1 (7) 7 (47)

PT

H 7 (47) 5 (33) 0 (0) 2 (13) 1 (7) 8 (53)

Tf M 7 (47) 2 (13) 3 (20) 3 (20) 0 (0) 8 (53)

CE

P 6 (40) 4 (27) 1 (7) 3 (20) 1 (7) 9 (60)

H 1 (7) 1 (7) 10 (67) 3 (20) 0 (0) 14 (93)

AC

Fn M 3 (20) 2 (13) 4 (27) 5 (33) 1 (7) 12 (80)

P 1 (7) 1 (7) 6 (40) 7 (47) 0 (0) 14 (93)

H: healthy, M: mucositis, P: peri-implantitis; Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td:
Treponema denticola, Tf: Tanerella forsythia, Fn: Fusobacterium nucleatum

0
 ACCEPTED MANUSCRIPT

Table 3: RT-PCR analysis results for the investigated potentially periodontal pathogenic
bacteria. Absolute numbers of samples with the corresponding detection limit is given first
and in brackets values are given as % of the 15 investigated samples. The prevalence shows
the general number of positive results in relation to the 15 total samples.

T
Concentrations Under Over detection limit prevalence

P
detection
limit

RI
0 - 0,2 >0,2 - 0,8 >0,8 - 1,4 >1,4 - 2,0 >2

Aa H 2 (13)

SC
13 (87) 2 (13) 0 (0) 0 (0) 0 (0)
M 9 (60) 2 (13) 2 (13) 1 (7) 1 (7) 6 (40)
P

NU
13 (87) 1 (7) 0 (0) 1 (7) 0 (0) 2 (13)
Pg H 10 (67) 3 (20) 2 (13) 0 (0) 0 (0) 5 (33)
M 8 (53)
MA
7 (47) 2 (13) 1 (7) 2 (13) 3 (20)
P 11 (73) 0 (0) 0 (0) 0 (0) 4 (27) 4 (27)
Td H 6 (40) 3 (20) 0 (0) 6 (40) 0 (0) 9 (60)
ED

M 7 (47) 2 (13) 2 (13) 3 (20) 1 (7) 8 (53)

P 7 (47) 3 (20) 0 (0) 4 (27) 1 (7) 8 (53)
PT

Tf H 7 (47) 5 (33) 0 (0) 1 (7) 2 (13) 8 (53)

M 6 (40) 4 (27) 4 (27) 1 (7) 0 (0) 9 (60)
CE

P 8 (53) 2 (13) 2 (13) 2 (13) 1 (7) 7 (47)

Fn H 0 (0) 0 (0) 1 (7) 13 (87) 1 (7) 15 (100)
AC

M 0 (0) 0 (0) 3 (20) 12 (80) 0 (0) 15 (100)

P 0 (0) 0 (0) 2 (13) 12 (80) 1 (7) 15 (100)
H: healthy, M: mucositis, P: peri-implantitis; Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td:
Treponema denticola, Tf: Tanerella forsythia, Fn: Fusobacterium nucleatum

0
 ACCEPTED MANUSCRIPT

Table 4: Comparison of PCR and RT-PCR. The prevalences of PCR and RT-PCR are
shown as absolute number and in brackets as % in relation to 45 investigated samples. The
exact concentration match means the exact match of PCR and RT-PCR. Kendalls Tau and

T
Spearman test results show the correlation between PCR and RT-PCR.

P
prevalence exact
prevalence Kendalls Tau

RI
bacteria RT-PCR concentration Spearman (R)
PCR [n(%)] (R)
[n(%)] match [n(%)]

SC
Aa 9 (20) 10 (22) 38 (84) 0.38 0.42

Pg 30 (67) 17 (38) 15 (33) 0.04 0.04

Td 15 (33)
NU
25 (55) 23 (51) 0.49 0.59
MA
Tf 25 (55) 24 (53) 25 (55) 0.65 0.78

Fn 40 (89) 45 (100) 13 (29) 0.49 0.61

ED

Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td: Treponema denticola, Tf: Tanerella forsythia,
Fn: Fusobacterium nucleatum
PT

Table 5: Results of sensitivity and specificity analysis for PCR and RT-PCR. Sensitivity and
CE

specificity for discrimination between healthy and mucositis as well as healthy and peri-
implantitis was determined for the five investigated bacteria.
AC

healthy vs mucositis healthy vs. peri-implantititis

method bacteria
sensitivity specificity sensitivity specificity
Aa 40% 87% 13% 87%
Pg 53% 67% 27% 67%
PCR Td 53% 40% 53% 40%
Tf 60% 47% 47% 47%
Fn 100% 100% 100% 100%
Aa 27% 87% 20% 87%
Pg 53% 20% 67% 20%
RT-PCR Td 27% 73% 47% 27%
Tf 53% 47% 60% 47%
Fn 80% 7% 93% 7%
Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td: Treponema denticola, Tf: Tanerella forsythia,
Fn: Fusobacterium nucleatum

0
 ACCEPTED MANUSCRIPT

Figure

Figure 1a: Prevalence of potentially periodontal pathogenic bacteria based on PCR in

healthy, mucositis and peri-implantitis. The p-values show the comparison of the bacterial
findings using PCR between the three peri-implant conditions and was determined using

T
Fisher-exact test (p<0.05)

P
RI
100 p = 0.41
90
p = 0.2

SC
80
70 p = 0.71
prevalence (%)

60

NU
p = 0.31
50 healthy
40 mucositis
p = 0.56
MA
30 peri-implantitis

20
10
ED

0
Aa Pg Td Tf Fn
investigated bacteria
PT

Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td: Treponema denticola, Tf: Tanerella forsythia,
Fn: Fusobacterium nucleatum
CE
AC

0
 ACCEPTED MANUSCRIPT

Figure 1b: Prevalence of potentially periodontal pathogenic bacteria by RT-PCR in healthy,

mucositis and peri-implantitis. The p-values show the comparison of the bacterial findings
using RT-PCR between the three peri-implant conditions and was determined using ANOVA
(p<0.05)

P T
p = 0.87
100

RI
90

SC
80
70
p = 0.92 p = 0.97
prevalence (%)

60 p = 0.16 healthy

NU
50 mucositis
p = 0.32
40 peri-implantitis
MA
30
20
10
ED

0
Aa Pg Td Tf Fn
investigated bacteria
PT

Aa: Aggregatibacter actinomycetemcomitans, Pg: Porphyromonas gingivalis, Td: Treponema denticola, Tf: Tanerella forsythia,
Fn: Fusobacterium nucleatum
CE
AC

0
 ACCEPTED MANUSCRIPT

Highlights

The investigated bacteria were found in healthy and diseased peri-implant conditions.
PCR and RT-PCR results were similar, without discrimination between conditions.
PCR-based test systems are sufficient for dental practice.
Benefit of detection of periodontal pathogens in peri-implantitis is questionable.

T
Clinical diagnoses of mucositis and peri-implantitis remain indispensable.

P
RI
SC
NU
MA
ED
PT
CE
AC