J Clin Periodontol 2013; 40: 1–7 doi: 10.1111/jcpe.

12021

Failure to detect an association Jamal M. Stein1,†, Sareh

Said Yekta1,†, Michael Kleines2,
Dilara Ok1, Adrian Kasaj3, Stefan

between aggressive periodontitis Reichert4, Susanne Schulz4 and

Simone Scheithauer5
1
Department of Operative Dentistry,

and the prevalence of Periodontology and Preventive Dentistry,

University Hospital (RWTH), Aachen,
Germany; 2Division of Virology, Department

herpesviruses of Hygiene, Microbiology, Social Medicine

Innsbruck Medical University, Innsbruck,
Austria; 3Department of Operative Dentistry
and Periodontology, University Medical
Center, Mainz, Germany; 4Department of
Stein JM, Said Yekta S, Kleines M, Ok D, Kasaj A, Reichert S, Schulz S, Scheithauer Operative Dentistry and Periodontology,
University School of Dental Medicine, Martin-
S. Failure to detect an association between aggressive periodontitis and the prevalence
Luther University, Halle-Wittenberg,
of herpesviruses. J Clin Periodontol 2013; 40: 1–7. doi: 10.1111/jcpe.12021.
Germany; 5Department of Infection Control
and Infectious Diseases, University Hospital
Abstract (RWTH), Aachen, Germany
Background: Herpes simplex virus type 1 (HSV-1), human cytomegalovirus
(HCMV) and Epstein–Barr virus (EBV) have been suspected to play a causal
role in periodontitis pathogenesis. The aim of this study was to determine the
prevalence of these viruses in subgingival plaque samples of Caucasian patients
with generalized aggressive periodontitis compared to periodontally healthy
controls.
Methods: A total of 65 patients with aggressive periodontitis and 65 unmatched
controls from Germany were investigated in the study. Subgingival plaque sam-
ples were analysed for the presence of HSV-1, EBV and HCMV by quantitative
real-time polymerase chain reaction assays. Viral antibody titres were determined
quantitatively by immunosorbent assays.
Results: DNA of HSV-1 and HCMV were detected in 1.5% of the patients and
controls, whereas EBV DNA was present in 10.8% and 13.9% respectively.
Detection rates of serum IgG against HSV-1 (76.1% versus 73.9%), EBV (98.5%
versus 96.9%), HCMV (47.7% versus 46.2%) and IgM levels against HSV-1
(6.2% versus 1.5%), EBV (0% versus 0%), HCMV (0% versus 1.5%) did not sig-
Key words: aggressive periodontitis;
nificantly differ between patients and controls. cytomegalovirus; Epstein–Barr virus; herpes
Conclusion: The data of our study do not suggest any contribution of HSV-1, simplex virus; pathogenesis
EBV or HCMV to aggressive periodontitis in a German population. Ethnic and
methodological aspects might have caused conflicting results of previous studies. Accepted for publication 9 September 2012

Periodontitis is an inflammatory dis- have been detected in subgingival pla-

Conflict of interest and source of
funding statement
The authors declare that they have no
conflict of interests. No external fund-
ing, apart from the support of the
author’s institution, was available for
this study.
†
These authors contributed equally to the
study.
© 2012 John Wiley & Sons A/S
ease that is known to have a primarily
(multi) bacterial aetiology (Kebschull
& Papapanou 2011, Wade 2011). In
the last 13 years, however, several
studies considered viruses to play a
role in the pathogenesis of gingivitis
and periodontitis. Mainly, herpesvi-
ruses such as herpes simplex virus
(HSV), Epstein–Barr virus (EBV) and
human cytomegalovirus (HCMV)
que samples of patients with chronic
(Contreras & Slots 2000, Ling et al.
2004, Bilichodmath et al. 2009)
and aggressive (Michalowicz et al.
2000, Saygun et al. 2004, Rotola et al.
2008) periodontitis, gingivitis (Imbro-
nito et al. 2008, Grenier et al. 2009)
and HIV-associated periodontitis
(Contreras et al. 2001, Grande et al.
2008, 2011). In most of these studies,
1
2 Stein et al.
the prevalence of herpesviruses was
increased compared with the peri-
odontitis-free control group. Other
investigations revealed a higher viral
load in disease-active or deep pockets
compared with non-active sites or
shallow pockets (Contreras & Slots
2000, Kamma & Slots 2003, Thoma-
sini et al. 2012). Based on these data, a
new pathogenesis model for periodon-
titis suggesting a bacterial–herpesviral
co-infection had been proposed (Slots
2005, 2010). Thereby, herpesviruses
(in particular EBV and HCMV) might
trigger disease progression in a local
immunosuppressive manner or have
direct cytopathic effects on fibroblasts
and immunocompetent cells (Contre-
ras & Slots 2000, Slots 2005, 2010).
However, maintenance of this
model is challenged by studies that
could not confirm an increased preva-
lence of herpesviruses (EBV, HCMV)
in chronic and aggressive periodonti-
tis (Dawson et al. 2009, Nibali et al.
2009) or that could not find differ-
ences in the viral load between shal-
low and deep pockets (Konstantinidis
et al. 2005). This inconsistency might
be caused by methodological reasons:
Most of the studies with positive
associations were done with a small
sample size (<30), limiting the degree
to which these results can be general-
ized. Furthermore, most of the previ-
ous findings were based on ultra-
sensitive molecular techniques (nested
PCR), which are known to carry the
risk of false-positive results due to
cross-contamination (Botero et al.
2008). Also, most of the studies were
performed in South American, Turk-
ish, Chinese, Africa Americans or in
ethnically mixed populations. As the
prevalence of herpesviruses depends
on the geographical, socioeconomic
and possibly the ethnical status (Nah-
mias et al. 1990, Pebody et al. 2004,
Hellenbrand et al. 2005) in various
populations, different associations
can be expected and, thus, findings
should be differentiated. For HSV-1,
the seroprevalence in Germany is
high, ranging from 85.2% (Western
part) to 88.5% (Eastern part). In
addition, seroprevalence is age depen-
dent, e.g. with a seroprevalence of
>60% in young German adults
compared to >80% in the age of
35–44 years and nearly 100% in the
elderly (Hellenbrand et al. 2005). A
very sophisticated investigation on
HCMV prevalence in the USA
revealed significant influence of age
(36% in childhood versus 91% in
those aged 80 years) and ethnicity
(51% in non-Hispanic, 76% in non-
Hispanic black people and 82% in
Mexican Americans) respectively
(Staras et al. 2006). The seropreva-
lence levels are in line with those
known from Germany (Enders et al.
2012), but lower than reported for
India and others (Kothari et al.
2002), and higher compared with his-
torical Canadian data (Embil et al.
1969). EBV prevalence reaches almost
90% at the age of 30–40 years in
Western/Central Europe compared
with a close to 100% positivity rate in
Central African populations. By age
of 4, virtually all people are infected
in developing countries, whereas sero-
prevalence in the United States
depends on the socio-economic status
(Vetsika & Callan 2004). Moreover,
for most of the herpesviruses, symp-
tomatic and even asymptomatic reac-
tivation is common, both leading to
shedding of viral DNA (Arora et al.
2010, Scheithauer et al. 2010). Hence,
detecting viral DNA is not equal to
implication of virus in pathogenesis.
As a consequence, for each condition,
the question is raised: bystander or
causative agent?
To the best of our knowledge, in
Central Europe, the prevalence of her-
pesviruses in periodontitis has not
been examined by now. Therefore,
and due to the contradictory results of
recent articles, it was the aim of our
study to investigate the prevalence of
herpesviruses HSV-1, EBV and
HCMV in periodontal pockets as well
as immunoglobulin G and M (IgG,
IgM) titres in German patients with
generalized aggressive periodontitis
(AP). To examine potentially aetiolo-
gic agents other than bacteria, we con-
sidered aggressive periodontitis as a
suitable disease model because early
onset, progressive course and frequent
disproportion between (often minimal)
microbial plaque and severe attach-
ment loss (Tonetti & Mombelli 1999)
makes this disease most likely to be
influenced by other aetiologic factors.
Materials and Methods
Study population
A total of 171 subjects, aged
23–70 years, were recruited for a
case-control study that was carried
out between May 2005 and December
2011 (Fig. 1). The patient group com-
prised 81 consecutive patients
referred to the Department of Opera-
tive Dentistry, Periodontology and
Preventive Dentistry, University
Hospital, Aachen, Germany with sus-
pected diagnosis of generalized
aggressive periodontitis (AgP). Con-
trols consisted of 90 unmatched
subjects considered free from peri-
odontitis that were enroled from the
same department. For evaluation of
anamnestic data, all participants were
interviewed according to a standard-
ized protocol. Subjects were asked
about their ethnic origin, medical
history, medications, smoking habit,
history of periodontitis and oral
hygiene. The study protocol was
approved by the Ethics Committee of
the University of Aachen, and written
informed consent was obtained from
all subjects before their examinations.
Criteria for exclusion from the
study were (i) non-Caucasian descent,
(ii) the presence of any systemic
diseases, (iii) intake of systemic anti-
biotics within the last 3 months, (iv)
the use of any oral antimicrobial
rinse devices in the last 3 months, (v)
drug-induced gingival hyperplasia,
(vi) chronic use of anti-inflammatory
drugs or (vii) current pregnancy.
Moreover, subjects were not included
if they had less than 20 teeth.
Patients and controls were assessed
in accordance with the classification
system of periodontal diseases (Armit-
age 1999). In particular, patients with
AgP were included in case of evidence
(from dental history and/or radio-
graphs) that the onset of the disease
occurred in the period before the age
of 35. They had to present a clinical
attachment loss of 5 mm or more on
at least 14 teeth. To exclude localized
aggressive periodontitis, patients had
to present at least three affected teeth
other than first molars or incisors.
Contrary to chronic periodontitis, the
severity of attachment loss was not
consistent with the amount of calcu-
lus, and more vertical than horizontal
approximal bone loss was evident in
the radiographs. Control subjects
were included if they were at least 30-
years old, had no or minimal signs of
gingival inflammation and they had
little or no attachment loss (probing
depth <3.5 mm and no gingival reces-
sion due to periodontitis [World
Health Organization 1978]). After
© 2012 John Wiley & Sons A/S

Selection of patients with AgP
(May 2005 - August 2011)
Potentially appropriate patients
with suspected diagnose AgP
(N = 80)
Exclusion of subjects
with non-Caucasian
descent (N = 5)
Caucasian patients with suspected
diagnose AgP (N = 75)
Exclusion of subjects
with systemic diseases
(N = 2)
Systemically healthy patients with
suspected diagnose AgP (N = 73)
Exclusion of subjects
not meeting inclusion
criteria for AgP after
periodontal exami-
nation (N = 8)
Systemically healthy patients with
AgP according inclusion and
exclusion criteria (N = 65)
(a)
Fig. 1. Flow diagram on the selection of patients with aggressive periodontitis (AgP)
(a) and controls (b) according to the inclusion and exclusion criteria.
exclusion of subjects not meeting the
inclusion criteria, the final study
cohorts consisted of 65 patients with
AgP and 65 controls. Fig. 1 shows a
flow diagram on selection of all cases
and controls according to the
described inclusion and exclusion cri-
teria.
Sample size of 65 cases and con-
trols had been considered acceptable
due to two issues: On the one hand,
we included more cases and controls
than in all previous studies on AgP
and herpesviruses (Yapar et al. 2003,
Saygun et al. 2004, Nibali et al.
2009). On the other hand, sample
size estimation (nQuery Advisor®
4.0, Statistical Solutions, Saugus,
MA, USA) revealed that Fisher’s
exact test with a 0.050 two-sided sig-
nificance level would have 80%
power in order to detect an odds
ratio (OR) of 3 or more between
two groups (for estimated 25% ver-
sus 50%) when the sample size in
each group is 65.
Clinical periodontal examination
The clinical periodontal examination
comprised the assessment of the
plaque index (Silness & Löe 1964),
gingival index (Löe & Silness 1963),
© 2012 John Wiley & Sons A/S
Aggressive periodontitis and viruses
Selection of control probands
(May 2005 - November 2011)
Potentially appropriate subjects
considered free from periodontitis
aged ≥30 (N = 91)
Exclusion of subjects
with non-Caucasian
descent (N = 7)
Caucasian subjects considered
free from periodontitis (N = 84)
Exclusion of subjects
with systemic diseases
(N = 3)
Systemically healthy subjects
considered free from periodontitis
(N = 81)
Exclusion of subjects
with periodontitis
(CAL≥3.5 mm) after
periodontal exami-
nation (N = 16)
Systemically healthy control
subjects without periodontitis
(N = 65)
(b)
probing depths (PD) and clinical
attachment level (CAL). PD and
CAL values were recorded at six
sites per tooth. The CAL (the dis-
tance between the cemento–enamel
junction and bottom of the pocket)
was obtained by adding the PD val-
ues to gingival recession values (the
distance between the gingival margin
and cemento–enamel junction). All
measurements were performed by
two examiners (JMS, DO) throughout
the study using a millimetre-graded,
pressure-sensitive probe (Hawe Click-
Probe, Kerr Hawe, Bioggio, Switzer-
land) set to a probing force of 0.25 N.
The examiner reliability was verified
by an inter-examiner calibration of
volunteers. Within a tolerance range
of ±1 mm, the kappa values for PD
and CAL measures ranged from 0.77
to 1.00, indicating a high level of
agreement.
Subgingival sample collection
Subgingival plaque samples of all
participants were taken from the
deepest pocket of each arch quad-
rant for the detection of herpesviral
nucleic acids (HSV-1, EBV, HCMV)
in the periodontal pockets. After
supragingival debridement, one ster-
3
ile paper point (Roeko, Langenau,
Germany) was inserted to the bot-
tom of each selected pocket for 15 s.
All four paper points with subgingi-
val plaque samples were pooled
together into a transfer tube that
was immediately transported to the
laboratory for further analyses. On
rare occasions in which laboratory
analysis could not be performed on
the same day, the tube was stored at
a temperature of 25°C and pro-
cessed on the next day.
DNA extraction and PCR procedures
DNA extraction was performed
using the QIAamp Virus Biorobot
9604 kit (Qiagen, Hilden, Germany).
The test assays applied were in-
house quantitative real-time PCR
assays using the LightCycler instru-
ment (Roche Diagnostics GmbH,
Mannheim, Germany) with analyti-
cal sensitivities from 10E2 (HCMV)
to 10E3 (HSV-1) geq/ml without any
cross-reactivity to other herpesviru-
ses (HSV-1: Genbank X14112, pos.
65644–65661 and 65916–65895
[primers], pos. 65771–65790 and
65796–65818 [probes]; EBV: Gen-
bank DQ279927, pos. 92907–92889
and 92721–92703 [primers], pos.
92828–92848 and 92853–92871
[probes]; HCMV: as described previ-
ously [Schaade et al. 2000]). Detec-
tion limits for PCR assays were
200 geq/ml (HCMV), 500 geq/ml
(EBV) and 1000 geq/ml (HSV-1)
respectively. All PCR analyses were
performed in a masked manner.
Quality control
Prior to our study, the applied sub-
gingival plaque sampling method
was tested for its ability to detect the
requested viral DNA. Therefore,
sterile paper points used for harvest-
ing subgingival plaque were spiked
with dilution series of HSV-1-,
HCMV- and EBV-DNA (control
plasmids harbouring the target
sequence of amplification). Paper tips
were eluted with 500 ll sterile water
in a lockable tube by vortexing. The
leachate was quantitatively trans-
ferred to the extraction step. DNA
was sufficiently eluated by re-sus-
pending. Detection limits were
defined and re-evaluated for paper
tips by dilution series. They proved
to be the same as for standard sam-
4 Stein et al.
ples as serum, cerebrospinal fluid,
urine, stool and other tissues (see
detection limits above). By a given
detection limit of 500 geq/ml (as an
example for EBV), the detectable
amount of virus in the eluted paper
tip was 250 copies. Including the
losses by the elution process, the
detection limit for the pooled four
paper tips resulted in 500 copies,
reflecting a mean of 125 copies for
each sample.
Furthermore, to compare paper
point sampling method with sampling
by gingival biopsies, in the first five
study participants both paper points
and gingival biopsies were harvested
from the deepest pocket of each arch
quadrant. In all test biopsies, identifi-
cation of the three herpesviruses was
identical with those gained by the
pooled paper points (EBV: one posi-
tive and four negative; HSV-1 and
HCMV: all negative). Thus, sampling
with paper points was considered suit-
able for this study.
Besides, for each PCR procedure,
positive (specific plasmids containing
the target region) and negative
(highly purified water) controls were
included according to the standards
of good laboratory practice.
Amplification was conducted with a
real-time assay using the Light
Cycler Instrument (Roche Diagnos-
tics GmbH, Mannheim, Germany).
The detection based on the fluores-
cence resonance energy transfer
(FRET) technology to allow not
only a sensitive amplification, but
also a high specificity. In contrast to
the basic (cheaper, easier and widely
used) technology using intercalating
substances, i.e. SYBR Green I, the
FRET technique is based on specific
probes, binding to the amplified
sequence in the case of a positive
result. Consequently, this method
outplays conventional assays and the
SYBR Green method due to a
higher specificity (Espy et al. 2006).
Quantification was performed by
standard curve methodology. Exter-
nal quality control was performed
by participating in round robin tests.
Antibody assays
Serum samples were collected from all
participants at the time of periodontal
examination. Serum was stored in a
20°C freezer up to several weeks
until tested. The seroprevalence for
HSV-1, EBV and Cytomegalovirus
(CMV) was determined by the Enzy-
gnost Anti-HSV/IgG/IgM, Anti-
CMV/IgG/IgM, Anti-EBV/IgG/IgM
(Siemens Healthcare Diagnostics Prod-
ucts, Marburg, Germany) representing
ELISA formatted assays. Results were
interpreted according to the manufac-
tures instructions. Undetermined
results (only for IgG anti-EBV) were
confirmed by immunofluorescent
assays and/or immunoblotting. Since
2008, HCMV antibody detection was
performed using the Architect technol-
ogy (Chemiluminiscence assays, Ab-
bott GmbH & Co.KG, Wiesbaden,
Germany). In accordance to PCR anal-
yses, serologic diagnosis was done in a
blinded manner.
Statistical analysis
Primary outcome values of continuous
variables were given as means and
standard deviation (SD). The unpaired
t-test for unequal variances was used
for comparison of these values between
the AgP group and controls. Compari-
son of categorical variables was per-
formed with the Fisher’s exact test.
The outcome variable was the direct
detection of the herpesviruses (HSV-1,
EBV, HCMV) and seropositivity of
IgG and IgM (IgG anti-HSV-1, IgM
anti-HSV-1, IgG anti-EBV, IgM anti-
EBV, IgG anti-HCMV, IgM anti-
HCMV). All statistical analyses were
performed in an explorative manner;
thus p-values of p < 0.05 can be inter-
preted as statistically significant test
results. Data processing and statistical
analyses were performed using soft-
Table 1. Demographic data and clinical periodontal parameters
AgP (N=65)
Age (years) 35.4 ± 5.9
Gender 39 (60.0%)
Females
Males 26 (40.0%)
Smoking 29 (44.6%)
Smokers
Pack-Years 6.3 ± 9.2
Periodontal parameters 0,9 ± 0,6
PI
GI 1.3 ± 0.7
PD (mm) 5.8 ± 1.2
CAL (mm) 6.7 ± 1.6
BOP 73.7 ± 32.6
N missing teeth 3.1 ± 3.9
*unpaired t-test
†
Fisher’s exact test
CAL, clinical attachment loss; GI, Gingival Index; PD, periodontal probing depth; PI,
Plaque Index.
ware packages (SAS, Version 9.2, SAS
Institute, Cary, NC, USA; Microsoft
Office Excel, Version 11.1.1, Red-
mond, WA, USA).
Results
The demographic and clinical peri-
odontal parameters are given in
Table 1. Patients with AgP were on
average younger and comprised
more smokers with a higher number
of pack years than among the
healthy control group. As expected,
mean values of all periodontal
parameters (plaque index, gingival
index, PD, CAL and BOP) were
significantly higher than in controls.
Also, patients had lost more teeth
than control subjects.
Subgingival plaque samples
revealed a very low prevalence of all
herpesviruses. Both, HSV 1 and
HCMV have been found in samples of
only one patient (1.5%) and one con-
trol (1.5%). Quantification resulted in
2.8 9 10E3 geq/ml (HSV-1) and 200
geq/ml (HCMV) in the AgP group as
well as 100 geq/ml (HSV-1) and
200 geq/ml (HCMV) in the control
group respectively. EBV was detect-
able in seven patients (10.8%) and
nine controls (13.9%). Positive
samples ranged from 350 geq/ml
to 1.2 9 10E4 geq/ml (median: 500
geq/ml) in AgP patients and from
150 geq/ml to 800 geq/ml (median:
500 geq/ml) in the controls. There
were no statistical differences between
the groups (Table 2).
The seroprevalence rates for
HSV-1 (76.4% versus 73.9%), EBV
Controls (N=65) p-value
40.0 ± 10.3 0.0024*
35 (53.9%) 0.5954†
30 (46.1%)
19 (29.2%) 0.1015†
2.4 ± 4.6 0.0026*
0,4 ± 0.4 < 0.0001*
0.5 ± 0.5 < 0.0001*
2.8 ± 0.3 < 0.0001*
3.1 ± 0.5 < 0.0001*
34.1 ± 22.1 < 0.0001*
1.2 ± 1.9 0.0011*
© 2012 John Wiley & Sons A/S

Table 2. Prevalence of herpesviral nucleic
acid in subgingival plaque samples (PCR)
AgP (N=65) Controls (N=65)
HSV-1 1 (1.5%) 1 (1.5%)
EBV 7 (10.8%) 9 (13.9%)
HCMV 1 (1.5%) 1 (1.5%)
EBV, Epstein–Barr virus; HCMV, human
cytomegalovirus; HSV-1, herpes simplex
virus 1.
For all comparisons: p > 0.05 (Fisher’s
exact test)
(98.5% versus 96.9%)and HCMV
(47.7% versus 46.2%) did not differ
significantly between AgP and con-
trols. Moreover, IgG seroprevalences
in our study were comparable with
published population data from Ger-
many (Table 3). IgM anti- EBV was
not detected in any sample, neither
in patients, nor in controls. Only one
control subject showed IgM anti-
HCMV. IgM anti-HSV-1 has been
detected in four patients and one
control subject (p > 0.05). In none of
the IgM anti-HSV-1-positive subjects
has HSV-1 been found in the peri-
odontal pockets. The missing signifi-
cance between the detection
frequencies of HSV-1, EBV and
HCMV (PCR and serum) in both
groups did not change when the
comparisons were done separately
for males and females as well as
smokers and non-smokers.
Discussion
In contrast to most of the previous
reports (Michalowicz et al. 2000,
Kamma & Slots 2003, Yapar et al.
2003, Saygun et al. 2004), the findings
of our present study do not suggest a
role of herpesviruses in aggressive
periodontitis within a central Euro-
Table 3. Seroprevalence (IgG, IgM) of HSV-1, EBV and HCMV
AgP (N=65)
IgG anti-HSV-1 49 (76.4%)
IgM anti-HSV-1 4 (6.2%)
IgG anti-EBV 64 (98.5%)
IgM anti-EBV 0 (0.0%)
IgG anti-HCMV 31 (47.7%)
IgM anti-HCMV 0 (0.0%)
*Published population data on seroprevalences in Germany.
† by the high risk of cross-contamina-
Hellenbrand et al. 2005
‡
Pottgiesser et al. 2006
§
Enders et al. 2012
For all comparisons between AgP and Controls: p > 0.05 (Fisher’s exact test)
AgP, Aggressive periodontitis; EBV, Epstein–Barr virus; HCMV, human cytomegalovirus;
HSV-1, herpes simplex virus 1.
© 2012 John Wiley & Sons A/S
Aggressive periodontitis and viruses
pean Caucasian population. There
was a very low prevalence of HSV-1
(1.5%) and HCMV (1.5%) as well as
a low detection rate of EBV (10.8%
versus 13.9%) in the periodontal
pockets of both patients with aggres-
sive periodontitis and healthy con-
trols. Also, seroprevalences (IgG,
IgM) of HSV-1, HCMV and EBV did
not significantly differ between cases
and controls, and IgG levels were
comparable to published population
data (Hellenbrand et al. 2005, Pott-
giesser et al. 2006, Enders et al.
2012). Although a high percentage of
the study participants had been
infected with HSV-1, EBV or HCMV
(IgG seropositivity), DNA of these
viruses was rarely present in subgingi-
val plaques. Even in IgM anti-HSV-1-
positive patients with AgP (6.2%),
HSV-1 DNA could not be detected in
the periodontal pockets.
Differences in the detection rate
of periodontal herpesviruses between
our study and many of the previous
reports might depend on different
reasons, including methodological
ones. In particular, the periodontal
disease studied, the applied viral
assays and ethnic or geographical
factors may in part explain the
observed discrepancies (Slots 2010).
First, in our study, we focused on
aggressive periodontitis only. Partic-
ular care was paid to the selection of
patients and controls. Only patients
with severe generalized periodontitis
with an age of onset  35 were
included. With a mean clinical
attachment level of 6.7 ± 1.6 mm,
disease severity was comparable or
even slightly increased compared
with those of AgP patients in other
studies (Yapar et al. 2003, Saygun
et al. 2004, Nibali et al. 2009). Also,
Controls (N=65) Population*
48 (73.9%) (85.2%–88.5%)†
1 (1.5%)
63 (96.9%) (83%–84%)‡
0 (0.0%)
30 (46.2%) (42.3%)§
1 (1.5%)
5
the mean probing depth of the sam-
ple sites (7.8 ± 1.9 mm) was not
below those of the cited previous
studies. Having a control group of
subjects that did not present any
CAL >3.5 mm due to periodontitis
and were older than 30 years of age,
a good contrast between cases and
controls could be achieved, which is
essential when disease-related aetio-
pathogenetic factors are studied.
Furthermore, to reduce the chance
of false (negative) results in the
detection of herpesviruses, we
excluded subjects who used local
antimicrobial rinsing solutions or
antibiotics within the last 3 months.
Nevertheless, the observed preva-
lence of the viruses was remarkably
low. Therefore, the selection criteria
for our study population could not
explain the contrast between our
results and the findings of most of
the previous studies.
Another more decisive reason for
the discrepancies of the results can be
seen in varying detection thresholds
and specificities of different PCR
techniques used to identify herpesvi-
ral nuclear acids. Most of the studies
that used highly sensitive methods
(nested PCR) found an increased
amount of herpesviruses and sug-
gested a contribution of these viruses
in the pathogenesis of periodontitis.
On the contrary, recent studies that
applied less sensitive, but more
specific methods as we used (identifi-
cation of 100 genome copies or more)
revealed lower prevalence data (Bote-
ro et al. 2008, Dawson et al. 2009,
Nibali et al. 2009). This might explain
why the low percentage of HCMV
found in our study population was in
good agreement to those reported by
Nibali et al. (2009) (0%) and Dawson
et al. (2009) (0.5%). Similarly, EBV
detection rate in our study was nearly
the same as in the control group of
Nibali et al. (2009) (10.3%), who had
even a decreased prevalence of EBV
in their patients with generalized AgP
(3.1%).
It should be noted that ultra-sensi-
tive PCR methods involve the risk of
overestimating the results (Botero
et al. 2008, Zhang et al. 2010) simply
tion within the assay procedure, as
well as by the technique-associated
lower specificity in comparison to
newer assay techniques. The latter is
even true within the group of auto-
6 Stein et al.
mated real-time PCR assays. It is well
known that the broadly used cheap
and easy-to-perform SYBR Green
detection method does not reveal the
high specificity as the FRET technol-
ogy does. Therefore, we can assume a
high level of true positive results in
our investigation. In other words,
given by the assay itself, the positivity
rate can be influenced. To achieve not
only specific but also robust and
therefore reproducible results, the
standard detection limit for clinical
samples was chosen. Moreover,
detecting DNA is not equal with
detecting a causative agent in disease
as shown particularly for HSV in
pneumonia (Scheithauer et al. 2010).
On the one hand, periodontal pockets
can be considered as ecological niche
of the oral cavity. We know that her-
pesviruses, in particular EBV and
CMV, can frequently be detected in
saliva (de França et al. 2011). There-
fore, it is not surprising that with
increasing depth of the periodontal
pockets,(i.e. increasing subgingival
niches of the oral cavity) there will
also be an increased probability of
detecting herpesviruses with highly
sensitive methods. Consequently, the
above cited studies noticed signifi-
cantly more herpesviruses in deep
pockets than in small pockets. How-
ever, this does not necessarily point to
a pathogenic role of these viruses, but
might rather indicate a coincidence of
viral nuclear acids in small
(“healthy”) versus large (“diseased”)
subgingival recesses of the oral cavity.
Moreover, in periodontal lesions, an
abundance of immunocompetent cells
can be found, e.g. neutrophils and
lymphocytes. As seroprevalence of
certain herpesviruses in the popula-
tion is very high (EBV: 80–100%),
and B lymphocytes (e.g. EBV) or T
lymphocytes (e.g. HCMV) can be
infected by herpesviruses, the detec-
tion of herpesviral nucleic acid in
periodontal pockets using highly sen-
sitive techniques in patients with
severe periodontitis could be an indi-
cation of the periodontal tissue break-
down, but does not need to point to a
causative relationship between herpe-
sviruses and periodontitis.
Finally, ethnic, geographical and
immunogenetic factors may also con-
tribute to explanation of the data.
Rotola et al. (2008) and Dawson
et al. (2009) emphasized the influence
of ethnic background to the results of
herpesvirus detection. To our knowl-
edge, this is the first study on this
topic in a German population. Possi-
bly, in the Mid-European population
herpesviruses HSV-1, HCMV and
EBV do not have the same prevalence
and impact on periodontal diseases.
Assumingly, other susceptibility fac-
tors such as the MHC polymorphism,
which plays an important role in
T-cell-dependent recognition of anti-
gens and also has an ethnically influ-
enced genetic background, could have
a similar influence as the proposed
pathogens and, therefore, cause dif-
ferences in immune response against
viral or bacterial agents (Stein et al.
2008). As viral antigen peptides are
presented by MHC class I molecules,
polymorphisms in the MHC class I
gene loci could influence susceptibility
against herpesvirus infections. MHC
class I antigens, HLA-A2 and -B5,
have been reported to be protective
against periodontitis, whereas HLA-A9
and -B15 seem to act as susceptibility
markers. According to the distribu-
tion of these markers within a popu-
lation, susceptibility to herpesviruses
could strongly be influenced by them,
whereby predominance of one or
more markers or combination of sev-
eral markers might have an additional
impact (Stein et al. 2003).
In conclusion, the data of our
study do not indicate any association
of herpesviruses, for Mid-European
patients, with aggressive periodonti-
tis. Although the previously proposed
model of a herpesviral–bacterial
co-infection (Slots 2010) represents a
plausible pathogenic concept, its
maintenance is compromised by the
growing inconsistency of the litera-
ture (present study, Nibali et al. 2009,
Dawson et al. 2009). Therefore, the
supposed causative role of these
viruses in periodontitis should be
questioned, and at least might be
referred not to all populations. Meth-
odological aspects, in particular
molecular methods with different
detection limits of virus genome cop-
ies, as wells as ethnic/geographical
descent and the influence of host
response factors might be reasons for
the discrepancies. Nevertheless, we
emphasize that the results of our
study can only be applied to a mid-
European Caucasian population and
the sample size is too small to general-
ize conclusions. In particular, with
the chosen sample size, associations
with odds ratios of 3 or more have
not been detected, whereas higher
sample sizes would have been neces-
sary to detect smaller effects.
We recommend that future studies
should verify our results in a larger
cohort with clearly characterized eth-
nic backgrounds. Thereby, standard-
ized and clinically relevant thresholds
for virus detection limits should be
defined to avoid overestimation (false
positive results) and allow compara-
bility with other populations.
References
Armitage, G. (1999) Development of a classifica-
tion system for periodontal diseases and condi-
tions. Annals of Periodontology 4, 1–6.
Arora, N., Novak, Z., Fowler, K.B., Boppana, S.
B. & Ross, S.A. (2010) Cytomegalovirus viruria
and DNAemia in healthy seropositive women.
Journal of Infectious Diseases 202, 1800–1803.
Bilichodmath, S., Mangalekar, S.B., Sharma, D.
C., Prabhakar, A.K., Reddy, S.B., Kalburgi,
N.B., Patil, S.R. & Bhat, K. (2009) Herpesvi-
ruses in chronic and aggressive periodontitis
patients in an Indian population. Journal of
Oral Science 51, 79–86.
Botero, J. E., Vidal, C., Contreras, A. & Parra, B.
(2008) Comparison of nested polymerase chain
reaction (PCR), real-time PCR and viral cul-
ture for the detection of cytomegalovirus in
subgingival samples. Oral Microbiology and
Immunology 23, 239–244.
Contreras, A., Mardirossian, A. & Slots, J. (2001)
Herpesviruses in HIVperiodontitis. Journal of
Clinical Periodontology 28, 96–102.
Contreras, A. & Slots, J. (2000) Herpesviruses in
human periodontal disease. Journal of Peri-
odontal Research 35, 3–16.
Dawson, D.R., Wang, C., Danaher, R.J., Lin, Y.,
Kryscio, R.J., Jacob, R.J. & Miller, C.S. (2009)
Real-time polymerase chain reaction to
determine the prevalence and copy number of
epstein-barr virus and cytomegalovirus DNA in
subgingival plaque at individual healthy and
periodontal disease sites. Journal of Periodon-
tology 80, 1133–1140.
Embil, J.A., Haldane, E.V., MacKenzie, R.A.E.
& van Rooyen, C.E. (1969) Prevalence of cyto-
megalovirus infection in a normal urban popu-
lation of Nova Scotia. Canadian Medical
Association Journal 101, 78–81.
Enders, G., Daiminger, A., Lindemann, L., Kno-
tek, F., Bäder, U., Exler, S. & Enders, M.
(2012) Cytomegalovirus (CMV) seroprevalence
in pregnant women, bone marrow donors and
adolescents in Germany, 1996-2010. Medical
Microbiology and Immunology, 201, 303–309.
Espy, M.J., Uhl, J.R., Sloan, L.M., Buckwalter,
S.P., Jones, M.F., Vetter, E.A., Yao, J.D.C.,
Wegenack, N.L., Rosenblatt, J.E., Cockerill, F.
R. & Smith, T.F. (2006) Real-Time PCR in
Clinical Microbiology: Applications for Rou-
tine Laboratory Testing. Clinical Microbiology
Reviews 19, 165–256.
de França, T.R., de Albuquerque, T.C.A., Gomes,
V.B., Gueiros, L.A., Porter, S.R. & Leao, J.C.
(2011) Salivary shedding of Epstein-Barr virus
and cytomegalovirus in people infected or not by
human immunodeficiency virus 1. Clinical Oral
Investigations 16, 659–664.
© 2012 John Wiley & Sons A/S

Grande, S.R., Imbronito, A.V., Okuda, O.S., Lot-
ufo, R.F., Magalhaes, M.H. & Nunes, F.D.
(2008) Herpes viruses in periodontal compro-
mised sites: comparison between HIV-positive
and –negative patients. Journal of Clinical Peri-
odontology 35, 838–845.
Grande, S.R., Imbronito, A.V., Okuda, O.S.,
Pannuti, C.M., Nunes, F.D. & Lima, L.A.
(2011) Relationship between herpesviruses and
periodontopathogens in patients with HIV and
periodontitis. Journal of Periodontology 82,
1442–1452.
Grenier, G., Gagnon, G. & Grenier, D. (2009)
Detection of herpetic viruses in gingival crevicu-
lar fluid of patients suffering from periodontal
diseases: prevalence and effect of treatment. Oral
Microbiology and Immunology 24, 506–509.
Hellenbrand, W., Thierfelder, W., Müller-Pebody,
B., Hamouda, O. & Breuer, T. (2005) Seropre-
valence of Herpes simplex virus Type 1 (HSV1)
and Type 2 (HSV2) in former East (FEG) and
West (FWG) Germany, 1997-1998. European
Journal of Clinical Microbiology & Infectious
Diseases 24, 131–135.
Imbronito, A.V., Okuda, O.S., Maria de Freitas,
N., Moreira Lotufo, R.F. & Nunes, F.D.
(2008) Detection of herpesviruses and peri-
odontal pathogens in subgingival plaque of
patients with chronic periodontitis, generalized
aggressive periodontitis, or gingivitis. Journal
of Periodontology 79, 2313–2321.
Kamma, J.J. & Slots, J. (2003) Herpesviralbacterial
interactions in aggressive periodontitis. Journal
of Clinical Periodontology 30, 420–426.
Kebschull, M. & Papapanou, P.N. (2011) Peri-
odontal microbial complexes associated with
specific cell and tissue responses. Journal of
Clinical Periodontology 38(Suppl. 11), 17–27.
Konstantinidis, A., Sakellari, D., Papa, A. &
Antoniadis, A. (2005) Real-time polymerase
chain reaction quantification of Epstein-Barr
virus in chronic periodontitis patients. Journal
of Periodontal Research 40, 294–298.
Kothari, A., Ramachandran, V.G., Gupta, P.,
Singh, B. & Talwar, V. (2002) Seroprevalence
of cytomegalovirus among voluntary blood
donors in Delhi, India. Journal of Health, Pop-
ulation and Nutrition 20, 348–351.
Ling, L.J., Ho, C.C., Wu, C.Y., Chen, Y.T. &
Hung, S.L. (2004) Association between human
herpesviruses and the severity of periodontitis.
Journal of Periodontology 75, 1479–1485.
Löe, H. & Silness, J. (1963) Periodontal disease in
pregnancy: prevalence and severity. Acta Odon-
tologica Scandinavica 21, 532.
Michalowicz, B. S., Ronderos, M., Camara-Silva,
R., Contreras, A. & Slots, J. (2000) Human
Clinical Relevance
Scientific rational for the study: Recent
studies have hypothesized a patho-
genic role of herpesviruses in peri-
odontitis.
© 2012 John Wiley & Sons A/S
Aggressive periodontitis and viruses
herpesviruses and Porphyromonas gingivalis
are associated with juvenile periodontitis. Jour-
nal of Periodontology 71, 981–988.
Nahmias, A.J., Lee, F.K. & Nahmias, B.S. (1990)
Sero-epidemiological and -sociological patterns
of herpes simplex infections in the world. Scandi-
navian Journal of Infectious Diseases 69, 19–36.
Nibali, L., Atkinson, C., Griffiths, P., Darbar, U.,
Rakmanee, T., Suvan, J. & Donos, N. (2009)
Low prevalence of subgingival viruses in peri-
odontitis patients. Journal of Clinical Periodon-
tology 36, 928–932.
Pebody, R.G., Andrews, N., Brown, D., Gopal,
R., de Melker, H., Francois, G., Gatcheva, N.,
Hellenbrand, W., Jokinen, S., Klavs, I., Kojou-
harova, M., Kortbeek, T., Kriz, B., Prosenc,
K., Roubalova, K., Teocharov, P., Thierfelder,
W., Valle, M., Van Damme, P. & Vranckx, R.
(2004) The seroepidemiology of herpes simplex
virus type 1and 2 in Europe. Sexually transmit-
ted infections 80, 185–191.
Pottgiesser, T., Wolfarth, B., Schumacher, Y.O. &
Bauer, G. (2006) Epstein–Barr Virus serostatus:
No difference despite aberrant patterns in ath-
letes and control group. Medicine and Science
in Sports and Exercise 38, 1782–1791.
Rotola, A., Cassai, E., Farina, R., Caselli, E.,
Gentili, V., Lazzarotto, T. & Trombelli, L.
(2008) Human herpesvirus 7, Epstein-Barr virus
and human cytomegalovirus in periodontal tis-
sues of periodontally diseased and healthy sub-
jects. Journal of Clinical Periodontology 35, 831
–837.
Saygun, I., Kubar, A., Ozdemir, A., Yapar, M. &
Slots, J. (2004) Herpesviral-bacterial interrela-
tionships in aggressive periodontitis. Journal of
Periodontal Research 39, 207–212.
Schaade, L., Kockelkorn, P., Ritter, K. & Kleines,
M. (2000) Detection of cytomegalovirus DNA in
human specimens by LightCycler PCR. Journal
of Clinical Microbiology 38, 4006–4009.
Scheithauer, S., Manemann, A. K., Krüger, S.,
Häusler, M., Krüttgen, A., Lemmen, S.W., Rit-
ter, K. & Kleines, M. (2010) Impact of herpes
simplex virus detection in respiratory specimens
of patients with suspected viral pneumonia.
Infection 38, 401–405.
Silness, J. & Löe, H. (1964) Periodontal disease in
pregnancy. Acta Odontologica Scandinavica 22,
121–135.
Slots, J. (2005) Herpesviruses in periodontal dis-
eases. Periodontology 2000 38, 33–62.
Slots, J. (2010) Herpesviral-bacterial interactions
in periodontal diseases. Periodontology 2000 52,
117–140.
Staras, S.A.S., Dollard, S.C., Radford, K.W.,
Flanders, D., Pass, R.F. & Cannon, M.J.
Principal findings: HSV-1, EBV and
HCMV were rarely found in German
patients with aggressive periodontitis.
There were no significant differences
between patients and controls.
7
(2006) Seroprevalence of Cytomegalovirus
Infection in the United States, 1988–1994. Clin-
ical Infectious Diseases 43, 1143–1151.
Stein, J., Reichert, S., Gautsch, A. & Machulla,
H.K. (2003) Are there HLA combinations typi-
cal supporting for or making resistant against
aggressive and/or chronic periodontitis? Journal
of Periodontal Research 38, 508–517.
Stein, J.M., Machulla, H.K., Smeets, R., Lampert,
F. & Reichert, S. (2008) Human leukocyte anti-
gen polymorphism in chronic and aggressive peri-
odontitis among Caucasians: a meta-analysis.
Journal of Clinical Periodontology 35, 183–192.
Thomasini, R.L., Bonon, S.H., Durante, P. & Costa,
S.C. (2012) Correlation of cytomegalovirus and
human herpesvirus 7 with CD3+ and CD3+ CD4
+ cells in chronic periodontitis patients. Journal of
Periodontal Research 47, 114–120.
Tonetti, M.S. & Mombelli, A. (1999) Early-onset
periodontitis. Annals of Periodontology 4, 39–53.
Vetsika, E.K. & Callan, M. (2004) Infectious
mononucleosis and Epstein-Barr virus. Expert
Reviews in Molecular Medicine 6, 1.
Wade, W.G. (2011) Has the use of molecular
methods for the characterization of the
humoral microbiome changed our understand-
ing of the role of bacteria in the pathogenesis
of periodontal disease? Journal of Clinical Peri-
odontology 38(Suppl. 11), 7–16.
World Health Organization. (1978) Epidemiology,
etiology, and prevention of periodontal dis-
eases. World Health Organization Technical
Report Series 621, 1–60.
Yapar, M., Saygun, I., Ozdemir, A., Kubar, A. &
Sahin, S. (2003) Prevalence of human herpesvi-
ruses in patients with aggressive periodontitis.
Journal of Periodontology 74, 1634–1640.
Zhang, S., Zhou, Y.H., Li, L. & Hu, Y. (2010)
Monitoring human cytomegalovirus infection
with nested PCR: comparison of positive rates
in plasma and leucocytes and with quantitative
PCR. Virology Journal 7, 73.
Address:
Jamal M. Stein
Department of Operative Dentistry
Periodontology and Preventive Dentistry
University Hospital Aachen (RWTH)
Pauwelsstrasse 30, D-52074 Aachen
Germany
E-mail: JStein@ukaachen.de
Practical implications: The proposed
role of herpesviruses in aggressive
periodontitis must be questioned.