J Clin Periodontol 2016; 43: 1003–1012 doi: 10.1111/jcpe.

12634

Does obesity influence the Suellen Silva Maciel1, Magda Feres1,

Tiago Eduardo Dias Gonc ! alves1,
Glaucia Santos Zimmermann1,2, He !lio

subgingival microbiota Doyle Pereira da Silva3, Luciene

Cristina Figueiredo1 and Poliana
Mendes Duarte1

composition in periodontal health 1

Department of Periodontology, Guarulhos
University, Sa~o Paulo, Brazil; 2Department of

and disease?
Dentistry, Federal University of Santa
Catarina, Floriano !polis, Santa Catarina,
Brazil; 3Biostatistics Unit, Public Health
Department, University of Barcelona,
Barcelona, Spain

Maciel SS, Feres M, Gonc!alves TED, Zimmermann GS, da Silva HDP, Figueiredo
LC, Duarte PM. Does obesity influence the subgingival microbiota composition in
periodontal health and disease?. J Clin Periodontol 2016; 43: 1003–1012. doi:
10.1111/jcpe.12634

Abstract
Aim: To evaluate whether obesity affects the subgingival microbial composition
of patients with periodontal health or chronic periodontitis (CP).
Materials and Methods: Based on periodontal parameters, body mass index and
waist-hip ratio, 166 patients were allocated into one of the following groups: Nor-
mal weight (NW) patients with periodontal health (n = 44), NW patients with CP
(n = 40), obese patients with periodontal health (n = 40) and obese patients
with CP (n = 42). Six subgingival biofilm samples per patient were analysed
for their content of 40 bacterial species using checkerboard DNA-DNA
hybridization.
Results: Obese patients with CP harboured higher levels and/or higher propor-
tions of several periodontal pathogens than those with NW and CP, including
Aggregatibacter actinomycetemcomitans, Eubacterium nodatum, Fusobacterium
nucleatum ss vincentii, Parvimonas micra, Prevotella intermedia, Tannerella for-
sythia, Prevotella melaninogenica and Treponema socranskii. The proportions of
most of these pathogens, as well Campylobacter rectus and Eikenella corrodens,
were more increased in the diseased sites of the obese patients than in those with
NW. Furthermore, the healthy sites of the obese patients, presenting or not CP,
also exhibited higher proportions of some of the pathogens than patients with Key words: chronic periodontitis; DNA
NW. probes; microbiology; obesity
Conclusions: Obesity is associated with increased levels and proportions of peri-
odontal pathogens, especially in patients with CP. Accepted for publication 3 October 2016

Conflict of interest and source of

Periodontitis is a destructive disease
funding statement resulting from an immunoinflamma-
The authors declare no conflicts of tory process in the periodontal tis-
interest with respect to the authorship sues triggered by pathogens present
and/or publication of this article. in the dental biofilm that may be
This study was supported by Sa ~o influenced by a variety of risk fac-
Paulo State Research Foundation tors. Obesity is recognized as one of
(FAPESP), Sa ~o Paulo, Sa ~o Paulo, the major risk factors for several
Brazil (#2010/01930-3; 2013/23767-5). medical conditions and has reached
epidemic proportions worldwide.
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Evidence has suggested a relation-
ship between periodontal diseases
and obesity. Cross-sectional and
prospective longitudinal studies and
systematic reviews have indicated
that obese patients exhibit greater
clinical attachment loss, patients
with periodontitis have increased
body mass index (BMI) and
patients presenting weight gain are
1003
1004 Maciel et al.
at higher risk of incidence of peri-
odontitis (Haffajee & Socransky
2009, Chaffee & Weston 2010,
Suvan et al. 2011, Nascimento et al.
2015).
The mechanisms underlying the
clinical associations between obesity
and periodontitis are not well under-
stood. Some evidence suggests that
immunoinflammatory processes may
provide a plausible pathomechanistic
link between both conditions. Taken
together, studies have shown that obese
patients with periodontitis present
higher levels of pro-inflammatory
biomarkers in their serum and gingival
crevicular fluid than non-obese patients
(Perri et al. 2012, Pradeep et al. 2013,
Zimmermann et al. 2013, Buduneli
et al. 2014).
Besides alterations in the imm-
unoinflammatory profile, differences
in the composition of the periodontal
microbiota may be another possible
explanation for the clinical associa-
tions observed between periodontal
diseases and obesity. This is an
important aspect to be scrutinized, as
it might directly interfere with pre-
ventive and treatment strategies.
Nonetheless, the relationship
between obesity and the oral micro-
biota has received little attention
(Haffajee & Socransky 2009, Mat-
sushita et al. 2015). Only one previ-
ous study has evaluated the
composition of the subgingival bio-
film in patients exhibiting different
levels of BMI and different periodon-
tal conditions (Haffajee & Socransky
2009). Out of the 40 species evalu-
ated in this study, only Tannerella
forsythia differed significantly among
groups. This periodontal pathogen
was in significantly higher propor-
tions in patients with BMI ≥ 30 kg/
m2 and periodontal health/gingivitis,
especially in patients younger than
46.8 years of age (Haffajee &
Socransky 2009). Therefore, due to
the scarce scientific evidence on
the topic and the relevance of the
information, the aim of this study
was to evaluate whether obesity
affects the subgingival microbial
composition of patients with peri-
odontal health or chronic periodonti-
tis (CP). This study tested the
hypothesis that obese patients har-
bour higher levels and proportions of
periodontal pathogens in the subgin-
gival biofilm than patients with nor-
mal weight.
Material and Methods
Sample size calculation
The ideal sample size to assure ade-
quate power for this study was cal-
culated considering a difference of at
least 1.8 percentage points (pp) for
the proportion of T. forsythia
between normal weight and obese
groups without periodontitis and, a
standard deviation of 2.2 pp (Haffa-
jee & Socransky 2009). Based on
these values, 32 patients per group
were defined as enough to provide
90% power with an a of 0.05.
Study population
Obese patients and patients with
normal weight, with or without CP,
were selected from the population
referred to the Periodontal Clinic of
Guarulhos University (Guarulhos,
SP, Brazil) and distributed into the
following groups: Normal weight
(NW) patients with periodontal
health (NWH) or CP (NWCP), and
obese patients with periodontal
health (ObH) or CP (ObCP). All eli-
gible individuals were invited to par-
ticipate in the study, thoroughly
informed of its nature, potential
risks and the benefits of their partici-
pation in the study and signed an
informed consent. They received
clinical and microbiological assess-
ment and were referred to the
University Dental Clinic in order to
receive periodontal treatment or pro-
phylaxis. The obese patients were
also referred to the Department of
Nutrition of the Guarulhos Univer-
sity in order to receive guidance
regarding their systemic health con-
dition. This study protocol was pre-
viously approved by the Guarulhos
University’s Ethics Committee in
Clinical Research (SISNEP 417-33/
2009).
Inclusion criteria
General
During the screening of volunteers,
detailed medical and dental histories,
as well as periodontal and anthropo-
metric measurements, were obtained.
All patients were >30 years old, had at
least 15 teeth, excluding third molars
and teeth with advanced decay. To be
included, all patients also had to pre-
sent glycated haemoglobin (HbA1c)
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
<6.5%, fasting plasma glucose (FPG)
ranging between 70 and 99 mg/dl and
C-reactive protein (CRP) levels
<6 mg/l. The levels of HbA1c (high-
performance liquid chromatography
method), FPG (glucose oxidase
method) and CRP (latex agglutination
test) were all assessed by the Guarul-
hos University Clinical Analysis Labo-
ratory during patient screening.
Obese patients
Obese patients had BMI ≥ 30 and
<40 kg/m2 and concurrent waist-hip
ratio (WHR) ≥ 0.85 for women and
WHR ≥ 0.90 for men (World Health
Organization 2008).
Patients with normal weight
Patients with normal weight pre-
sented BMI ranging from 18.5 to
24.9 kg/m2 and concurrent
WHR < 0.85 for women and
WHR < 0.90 for men (World Health
Organization 2008).
Patients with generalized CP
Patients with generalized CP were
defined as having >30% of sites with
concomitant probing depth (PD)
and clinical attachment level
(CAL) ≥ 4 mm and bleeding on
probing (BoP) (Armitage 1999), and
a minimum of six teeth distributed
in the different quadrants presenting
at least one site with PD and
CAL ≥ 5 mm and BoP.
Patients with periodontal health
Patients with no history of periodon-
titis and no sites with PD and CAL
>3 mm concomitantly.
Exclusion criteria
Exclusion criteria were pregnancy,
lactation, current smoking and
smoking within the past 10 years,
subgingival periodontal therapy
including non-surgical and surgical
scaling and root planing during the
previous 12 months, use of antibi-
otic, anti-inflammatory, immunosup-
pressive and lipid-lowering (e.g.
statins) therapies during the previous
6 months, regular use of mouthrinses
containing antimicrobials, use of
orthodontic appliances, presence of
systemic conditions that could affect
the progression of periodontitis and/
or gain/loss of weight (e.g. diabetes
mellitus [DM], immunological disor-
ders, osteoporosis, hypothyroidism,

hyperthyroidism) and history of car-
diovascular diseases (e.g. coronary
artery diseases, angina, myocardial
infarction, stroke). Patients that were
underweight (i.e. BMI < 18.5 kg/
m2), overweight (i.e. BMI ≥ 25 and
<29.9 kg/m2) or exhibiting extreme
obesity (i.e. BMI ≥ 40 kg/m2) were
also excluded.
Periodontal measurements and calibration
exercise
One examiner performed (S.S.M) all
clinical examinations. The following
parameters were assessed at six sites
(mesio-buccal, mid-buccal, disto-buc-
cal, mesio-lingual, mid-lingual, disto-
lingual) per tooth, excluding third
molars, using a manual periodontal
probe (UNC15, Hu-Friedy, Chicago,
IL, USA): visible plaque accumulation
(1/0), BoP (1/0), suppuration (1/0),
PD (mm) and CAL (mm). Calibration
was conducted according to the proto-
col developed by Araujo et al. (2003),
and the standard error of measure-
ment (SE) was calculated. The exam-
iner participated in a calibration
exercise and examined one quadrant
with at least six teeth of 10 non-study
patients with CP. Initially, the exam-
iner measured PD and CAL in a given
quadrant and 60 minutes later, this
same procedure was repeated. There-
fore, all 10 patients were probed twice
in the same visit by the examiner.
Upon completion of all measure-
ments, the intra-examiner variabilities
for PD and CAL measurements were
assessed. The intra-examiner variabil-
ity was 0.21 mm for PD and 0.24 mm
for CAL. The agreement for categori-
cal variables [e.g. BoP and plaque] was
92% (Kappa-light test).
Anthropometric measurements
One trained examiner (T.E.D.G.) per-
formed all anthropometric measure-
ments, including weight (kg), height
(m), waist (cm) and hip circumferences
(cm). BMI was calculated as the
weight divided by the square of height
(kg/m2). The waist-hip ratio (WHR)
was calculated as the ratio of waist to
hip circumference.
Microbiological analysis
After supragingival plaque removal,
subgingival biofilm samples were
collected with individual sterile
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Subgingival microbial profile in obesity
mini-Gracey curettes from six non-
contiguous sites per subject dis-
tributed in different quadrants, as
follows: Patients with CP – three
sites with PD ≤ 3 mm with no BoP
(i.e. healthy sites) and three sites
with PD and CAL >4 mm with
BoP (i.e. diseased sites); Patients
without CP – six sites with
PD ≤ 3 mm with no BoP (i.e.
healthy sites). The samples were
immediately placed in separate
microtubes containing 0.15 ml of
TE (10 mM Tris-HCl, 1 mM
EDTA, pH 7.6) and 100 ll of
0.5 M NaOH were added to each
tube. Subsequently, the samples
were evaluated for their content of
40 bacterial species by the checker-
board DNA-DNA hybridization
technique, as previously described
(Socransky et al. 1994, Mestnik
et al. 2010).
Statistical analysis
The percentages of sites with plaque
accumulation, BoP and PD ≥ 5 mm,
the mean full-mouth PD and CAL,
the mean levels and proportions of
the 40 bacterial species, the mean
proportions of the microbial complex
(Socransky et al. 1998) and the
anthropometric parameters were
computed for each individual and
subsequently, across groups. The
mean proportions of each bacterial
species at healthy sites and diseased
sites were determined separately for
each subject and then across each
group. Data were examined for nor-
mality by Shapiro–Wilk test and
non-parametric methods were used
for data that did not achieve normal
distribution. The significance of dif-
ferences in age, plaque accumulation,
PD, CAL, BMI and WHR was com-
pared using ANOVA, followed by the
Bonferroni test. The significance of
differences among groups for BoP
was compared using the Kruskal–
Wallis test, followed by the Dunn
test. The Mann–Whitney test was
used to compare the percentage of
sites with PD ≥ 5 mm and the CAL
and PD means of the sampled sites
between NWCP and ObCP groups.
The significance of difference
between groups for gender was com-
pared by the Chi-square test. Micro-
biological data were first transformed
using a Box-Cox to correct for asym-
metry, generate normal distribution
1005
of the data and stabilize variance.
Subsequently, microbiological data
were analysed using mixed-model
ANOVA for the comparison of levels
and proportions of individual bacte-
ria species and proportions of micro-
bial complexes with adjustments for
age, using individuals as the random
effect and obesity and/or periodonti-
tis in the shallow and/or deep sites as
fixed effects (SAS PROC MIXED,
SAS 9.1.2, SAS Institute Inc., Cary,
NC, USA). Afterwards, post hoc
analyses with the Bonferroni correc-
tion were performed. The residual
analysis validated the assumed mod-
els. Correlations between the propor-
tions of the bacterial species of the
red and orange complexes and the
anthropometric parameters were per-
formed using the Pearson Correla-
tion. The level of significance was set
at 5%.
Results
This study was conducted between
July 2011 and December 2014. One
hundred and sixty-six patients rang-
ing from 31 to 65 years of age were
selected out of the 910 screened. The
final numbers of patients in each
group were as follows: NWH;
n = 44, NWCP; n = 40, ObH; n = 40
and ObCP; n = 42.
Clinical and demographic characteristics
There were no differences among
groups for gender distribution
(p > 0.05). The mean age in the
ObCP group was higher than in the
NWH and ObH groups (p < 0.05).
All periodontal parameters were
higher in the NWCP and ObCP
groups than in the NWH and ObH
groups (p < 0.05). Furthermore,
BMI and WHR were higher in the
two groups of obese patients than in
the NW groups (p < 0.05; Table 1).
Microbiological results
Figure 1 presents the mean propor-
tions of the microbial complexes for
each group. The proportions of the
red and orange complexes were sig-
nificantly higher in both groups with
CP than in the two periodontally
healthy groups (p < 0.05).
Figure 2 presents the mean levels
(Fig. 2A) and proportions (Fig. 2B)
1006 Maciel et al.

Table 1. Demographic, periodontal (full-mouth and sampled sites) and anthropometric parameters for all groups (mean ! SD)
Parameters Groups

NWH (n = 44) NWCP (n = 40) ObH (n = 40) ObCP (n = 42) p-values

Gender (% female) 79.5 57.5 73.8 69.2 0.16*

Age (years) 45.2 ! 8.0a 47.80 ! 7.8 44.5 ! 8.2a 50.13 ! 6.6b 0.005#
Plaque (%) 22.4 ! 13.1a 68.55 ! 27.5b 26.1 ! 15.1a 64.91 ! 25.1b <0.0001#
PD (mm)
Full-mouth 2.2 ! 0.16a 3.4 ! 0.7b 2.3 ! 0.1a 3.5 ! 0.5b <0.0001#
Healthy sampled sites 2.3 ! 0.11 2.7 ! 0.3 2.4 ! 0.4 2.8 ! 0.3 0.23#
Diseased sampled sites – 6.6 ! 1.7 – 6.8 ! 1.6 0.18**
CAL (mm)
Full-mouth 2.3 ! 0.2a 4.2 ! 1.0b 2.3 ! 0.1a 4.3 ! 1.0b <0.0001#
Healthy sampled sites 2.3 ! 0.3 3.0 ! 0.4 2.4 ! 0.5 3.1 ! 0.3 0.18#
Diseased sampled sites – 7.5 ! 2.0 – 7.5 ! 2.2 0.13**
BoP (%) 14.1 ! 15.4a 61.2 ! 23.3b 9.2 ! 1.2a 59.2 ! 22.3b <0.0001‡
Sites with PD ≥ 5 mm (%) – 23.2 ! 13.3 – 25.2 ! 13.6 0.31**
BMI (kg/m2) 23.2 ! 1.9a 23.2 ! 1.6a 32.9 ! 3.6b 33.7 ! 2.7b <0.0001#
WHR 0.81 ! 0.05a 0.82 ! 0.05a 0.92 ! 0.1b 0.97 ! 0.06b <0.0001#

BoP, bleeding on probing; BMI, body mass index; CAL, clinical attachment level; PD, probing depth; WHI, waist-hip ratio.
NWH: individuals with normal weight and periodontal health; NWCP: individuals with normal weight and chronic periodontitis (CP);
ObH: obese patients with periodontal health; ObCP: obese patients with CP.
*Chi-square test.
#
ANOVA and Bonferroni test.
‡
Kruskal–Wallis and Dunn tests.
**Mann–Whitney test.
Different letters indicate differences among groups (p < 0.05).

Fig. 1. Mean proportions of the microbial complexes for each group. The colours represent the different microbial complexes, accord-
ing to Socransky et al. 1998. The grey colour represents species that did not fall into any complex (“others”), and the Actinomyces clus-
ter is represented in blue. Different letters indicate significant differences among groups (p < 0.05) by mixed-model ANOVA, followed by
post hoc analyses with the Bonferroni correction. NWH: individuals with normal weight and periodontal health; NWCP: individuals
with normal weight and chronic periodontitis (CP); ObH: obese patients with periodontal health; ObCP: obese patients with CP.

of the 40 bacterial species studied
for the NWCP and ObCP groups,
considering all sampled sites (i.e.
healthy and diseased sites). The
ObCP group presented higher
counts and proportions of one spe-
cies from the purple complex (Veil-
lonella parvula), one species from
the yellow complex (Streptococcus
gordonii,), one species from the
green complex (Aggregatibacter acti-
nomycetemcomitans), four species
from the orange complex (Eubac-
terium nodatum, Fusobacterium
nucleatum ss vincentii, Parvimonas
micra and Prevotella intermedia)
and two species from the “others”
group (Streptococcus anginosus and
Treponema socranskii) than the
NWCP group (p < 0.05). Further-
more, the mean counts of Actino-
myces gerencseriae, Streptococcus
intermedius, Streptococcus mitis,
Campylobacter gracilis, T. forsythia
and Prevotella melaninogenica were
also higher in the ObCP than in the
NWCP group (p < 0.05).
Figure 3 presents the mean pro-
portions of the 40 bacterial species
studied for the ObCP and NWCP
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
groups, considering only the healthy
sites (PD ≤ 3 mm without BoP)
(Fig. 3A) and only the diseased sites
(PD >4 mm with BoP) (Fig. 3B).
The proportions of P. intermedia
and T. socranskii were higher while
the proportions of Streptococcus ora-
lis and Streptococcus sanguinis were
lower in the healthy sites of the
ObCP group, when compared to
those of the NWCP group
(p < 0.05). Furthermore, the diseased
sites of the ObCP group demon-
strated significantly higher propor-
tions of 13 bacterial species than
 Subgingival microbial profile in obesity 1007

Fig. 2. Mean levels (A) and proportions (B) of the 40 bacterial species studied for the NWCP and ObCP groups, considering all
sampled sites (i.e. healthy and diseased sites). The species were ordered according to the microbial complexes described by Socran-
sky et al. (1998). *Significant differences among groups by mixed-model ANOVA and Bonferroni correction (p < 0.05). NWCP: indi-
viduals with normal weight and chronic periodontitis (CP); ObCP: obese patients with CP.

© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
1008 Maciel et al.
those of the NWCP group, including
one species from the Actinomyces
cluster (Actinomyces israelli), one
species from the purple complex
(V. parvula), one species from the
yellow complex (S. mitis), two spe-
cies from the green complex (A. acti-
nomycetemcomitans and Eikenella
corrodens), three species from the
orange complex (Campylobacter rec-
tus, E. nodatum and F. nucleatum ss
vincentii), one species from the red
complex (T. forsythia), and four spe-
cies from the “others” group
(Leptotrichia buccalis, P. melanino-
genica, S. anginosus and T. socran-
skii) (p < 0.05). On the other
hand, the diseased sites of the
NWCP group harboured higher
proportion of A. gerencseriae
than those of the ObCP group
(p < 0.05).
Figure 4 presents the mean levels
(Fig. 4A) and proportions (Fig. 4B)
of the 40 bacterial species studied
for the NWH and ObH groups. The
mean levels of Actinomyces naes-
lundii and Actinomyces oris (Actino-
myces cluster) were higher in the
NWH group than in the ObH group
(p < 0.05). Furthermore, the ObH
group exhibited higher mean propor-
tions of Actinomyces odontolyticus
(purple complex), S. gordonii, (yel-
low complex) and Fusobac-
terium nucleatum ss polymorphum
(orange complex) than the NWH
(p < 0.05).
Correlations
The proportions of E. nodatum,
P. intermedia and Streptococcus con-
stellatus positively correlated with
WHR in the NWCP group
(p < 0.05). The proportion of
T. forsythia presented positive corre-
lations with BMI and WHR and the
proportion of Porphyromonas gingi-
valis presented a positive correlation
with WHR in the ObH group
(p < 0.05). The proportions of
F. polymorphum and Fusobacterium
periodonticum positively correlated
with BMI in the ObCP group
(p < 0.05; Table 2).
Discussion
The overall results of this study
demonstrated that the subgingival
biofilm of obese patients with CP
harboured higher levels and/or
proportions of several periodontal
pathogens than that of patients with
CP and normal weight, including
A. actinomycetemcomitans, E. noda-
tum, F. nucleatum ss vincentii, P. mi-
cra, P. intermedia, T. forsythia,
P. melaninogenica and T. socranskii.
The proportions of most of these
pathogens, as well as others such as
C. rectus and E. corrodens were
higher in the diseased sites of the
obese patients than the diseased sites
of patients with normal weight.
Remarkably, the healthy sites of the
obese patients, presenting or not CP,
were colonized by higher propor-
tions of some pathogens (e.g. P. in-
termedia, T. socranskii and
F. nucleatum ss polymorphum) when
compared to those of patients with
normal weight. In addition, the pro-
portions of certain pathogens from
the orange and red complexes posi-
tively correlated with the anthropo-
metric parameters in the NWCP,
ObH and ObCP groups. Taken
together, these findings suggest that
obesity, a very common systemic
condition, might affect the composi-
tion of the subgingival microbiota at
shallow and deep sites by increasing
the levels and proportions of some
pathogenic species, which might, in
turn, enhance the risk of periodonti-
tis development and progression.
One could argue that these microbi-
ological findings may be attributed
to a more severe periodontitis in the
sampled sites of the obese patients.
However, it is important to observe
that obese and normal weight
patients were matched for the sever-
ity of periodontitis at sampled site
and full-mouth levels.
Only one previous study also
investigated the possible impact of
obesity on the subgingival micro-
biota, comparing the proportions of
these same 40 bacteria species
among patients with healthy/gingivi-
tis and periodontitis presenting dif-
ferent categories of BMI (Haffajee &
Socransky 2009). The only statisti-
cally significant difference observed
among groups was for the propor-
tion of T. forsythia, which was
greater in patients presenting
BMI ≥ 30 kg/m2 and periodontal
health/gingivitis than in patients pre-
senting normal weight or overweight
and periodontal health/gingivitis
(Haffajee & Socransky 2009). In this
study, the obese patients with
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
periodontal health exhibited a higher
proportion of F. nucleatum ss poly-
morphum, an orange complex patho-
gen, than the periodontally healthy
patients with normal weight. Fur-
thermore, in the healthy sites of
patients with CP, the proportions of
the pathogenic species, P. intermedia
and T. socranskii, were higher in
obese patients than in individuals
with normal weight. Notably, the
proportions of T. forsythia presented
moderate and strong positive corre-
lations with BMI and WHR, respec-
tively, in patients presenting
periodontal health and obesity.
Divergences between this study and
the study of Haffajee & Socransky
(2009) might be attributed to the dif-
ferences in the experimental groups,
methods and data analysis. Firstly,
the abovementioned study selected
participants from a series of ran-
domized clinical trials and did not
select the participants specifically to
examine the relationship between
periodontal status and obesity. Fur-
thermore, the authors included over-
weight patients in their multiple
statistical analyses, used only BMI
as the anthropometric parameter to
define obesity and entered patients
with minor and localized periodonti-
tis. Finally, they did not exclude
smokers and did not adjust for the
smoking status in the microbiologi-
cal analyses. Despite all these
methodological differences, the cur-
rent and the abovementioned study
supports the notion of the possible
role of pathogens in shallow sulci/
pockets of the obese patients.
The actual mechanisms underlying
the findings of this study, suggesting
that obesity is related to increased
levels of pathogens in the subgingival
microbiota (especially in patients with
CP), are unknown. In the medical
field, obesity has been linked to
patients’ increased susceptibility to
developing a variety of infections,
probably due to the dysregulation in
immune function and consequently,
in pathogen defense (Sanz & Moya-
P!erez 2014, Kaspersen et al. 2015,
Tagliabue et al. 2016). Evidence has
highlighted a substantial impact of
obesity on the disruption of lymphoid
tissue integrity, alterations in leuco-
cyte development, phenotypes, and
activity and on the general coordina-
tion of innate and adaptive immune
responses. All these immunological
 Subgingival microbial profile in obesity 1009

Fig. 3. Mean proportions of DNA probes for 40 subgingival species in the healthy (PD ≤ 3 mm without bleeding on probing, BoP)
(A) and diseased sites (PD > 4 mm with BoP) (B) for ObCP than in NWCP groups. The species were ordered according to the
microbial complexes described by Socransky et al. (1998). *Significant differences among groups by mixed-model ANOVA and Bonfer-
roni correction (p < 0.05). NWCP: individuals with normal weight and chronic periodontitis (CP); ObCP: obese patients with CP.

© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
1010 Maciel et al.

Fig. 4. Mean levels (A) and proportions (B) of the 40 bacterial species studied for the NWH and ObH groups. The species were
ordered according to the microbial complexes described by Socransky et al. (1998). *Significant differences among groups by
mixed-model ANOVA and Bonferroni correction (p < 0.05). NWH: individuals with normal weight and periodontal health; ObH:
obese patients with periodontal health.

© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
 Subgingival microbial profile in obesity 1011

Table 2. Pearson correlation coefficients for the bacterial species of the red and orange complexes and the anthropometric parameters, con-
sidering all sampled sites
Complex Species NWH NWCP ObH ObCP

BMI WHR BMI WHR BMI WHR BMI WHR

Orange Campylobacter gracilis 0.07 "0.16 0.08 0.23 0.0006 0.08 "0.10 "0.13
Campylobacter rectus 0.03 "0.04 "0.08 0.20 "0.05 0.04 "0.29 "0.05
Campylobacter showae 0.02 0.15 "0.17 0.15 0.15 0.24 "0.30 "0.22
Eubacterium nodatum "0.11 0.03 0.02 0.42* "0.15 "0.06 0.09 "0.11
Fusobacterium nuc ss nucleatum "0.06 0.005 "0.13 0.23 "0.16 "0.08 "0.11 "0.27
Fusobacterium nuc ss polymorphum 0.03 0.03 "0.17 0.05 "0.09 "0.07 0.33* "0.22
Fusobacterium nuc ss vincentii "0.10 0.15 "0.13 0.06 "0.12 0.08 "0.07 "0.17
Fusobacterium periodonticum "0.04 0.11 "0.18 0.16 0.07 0.14 0.42* "0.06
Parvimonas micra "0.13 0.10 "0.15 0.13 "0.004 0.06 "0.01 "0.22
Prevotella intermedia "0.11 0.23 "0.15 0.39* "0.02 "0.11 0.08 0.22
Prevotella nigrescens "0.30 "0.02 "0.29 0.10 "0.06 "0.02 0.14 0.21
Streptococcus constellatus 0.04 0.33 0.01 0.31* 0.06 "0.07 "0.02 "0.14
Red Tannerella forsythia "0.22 "0.11 "0.14 0.20 0.33* 0.74* "0.22 "0.14
Porphyromonas gingivalis "0.05 0.08 "0.22 0.03 0.21 0.38* "0.03 0.25
Treponema denticola 0.02 0.19 "0.19 0.25 0.009 "0.12 "0.24 0.01

NWH: individuals with normal weight and periodontal health; NWCP: individuals with normal weight and chronic periodontitis (CP);
ObH: obese patients with periodontal health; ObCP: obese patients with CP.
*Pearson Correlation; p < 0.05.

changes have been associated with an
overall negative impact of obesity on
chronic disease progression and
defence against infection (Sanz &
Moya-P!erez 2014, Tagliabue et al.
2016). Therefore, we speculate that,
as in other infections, obesity-asso-
ciated metabolic/immune changes in
periodontal tissues (Perri et al. 2012,
Pradeep et al. 2013, Zimmermann
et al. 2013, Buduneli et al. 2014)
might affect the subgingival microbial
colonization profile by favouring the
overgrowth of certain pathogenic spe-
cies, due to alterations in the pocket/
sulcus environment and/or periodon-
tal tissue host defences. The obesity-
induced alterations in the immunoin-
flammatory system, coupled with
changes in the microbiological pro-
file, might ultimately worsen the risk
of development, the extent and the
progression of periodontitis. Future
studies assessing, jointly, the microbi-
ological and immunoinflammatory
aspects of the periodontal sites of
obese patients would be of impor-
tance to confirm this hypothesis.
The main strengths of this study
are the exclusion of critical confound-
ing factors for periodontitis and obe-
sity, such as smoking, DM and high
levels of CRP, as well as the control
for the possible impact of age by
using statistical adjustments. In addi-
tion, the status of obesity was defined
based on BMI and WHR together,
to provide information regarding
body fat distribution, abdominal obe-
sity and visceral fat mass, which are
more closely associated to obesity-
related complications. This study also
has some limitations. Firstly, the find-
ings cannot be generalized to patients
with localized and mild periodontitis,
or to patients with overweight or
extreme obesity. Furthermore, the
cross-sectional design of this study
does not allow the establishment of
causal inferences and the temporal
association between the obesity and
the composition of the subgingival
microbiota, or the long-term clinical
consequences of such a relationship.
In conclusion, the results of this
study suggest that obesity interferes
with the subgingival microbial com-
position, as this systemic condition is
related to increased levels and pro-
portions of periodontal pathogens,
mainly in patients with CP.
References
Araujo, M. W., Hovey, K. M., Benedek, J. R.,
Grossi, S. G., Dorn, J., Wactawski-Wende, J.,
Genco, R. J. & Trevisan, M. (2003) Repro-
ducibility of probing depth measurement using
a constant-force electronic probe: analysis of
inter- and intraexaminer variability. Journal of
Periodontology 74, 1736–1740.
Armitage, G. C. (1999) Development of a classifi-
cation system for periodontal diseases and con-
ditions. Annals of Periodontology 4, 1–6.
Buduneli, N., Bıyıko"glu, B., Ilgenli, T., Buduneli,
E., Nalbantsoy, A., Sarac!, F. & Kinane, D. F.
(2014) Is obesity a possible modifier of peri-
odontal disease as a chronic inflammatory
process? A case-control study. Journal of Peri-
odontal Research 49, 465–471.
Chaffee, B. W. & Weston, S. J. (2010) Association
between chronic periodontal disease and obe-
sity: a systematic review and meta-analysis.
Journal of Periodontology 81, 1708–1724.
Haffajee, A. D. & Socransky, S. S. (2009) Rela-
tion of body mass index, periodontitis and
Tannerella forsythia. Journal of Clinical Peri-
odontology 36, 89–99.
Kaspersen, K. A., Pedersen, O. B., Petersen, M.
S., Hjalgrim, H., Rostgaard, K., Møller, B. K.,
Juul-Sørensen, C., Kotz!e, S., Dinh, K. M.,
Erikstrup, L. T., Sørensen, E., Thørner, L. W.,
Burgdorf, K. S., Ullum, H. & Erikstrup, C.
(2015) Obesity and risk of infection: results
from the Danish Blood Donor Study. Epidemi-
ology 26, 580–589.
Matsushita, K., Hamaguchi, M., Hashimoto, M.,
Yamazaki, M., Yamazaki, T., Asai, K.,
Yamori, M., Bessho, K., Toda, H., Hasegawa,
G., Nakamura, N. & Fukui, M. (2015) The
novel association between red complex of oral
microbe and body mass index in healthy Japa-
nese: a population based cross-sectional study.
Journal of Clinical Biochemistry and Nutrition
57, 135–139.
Mestnik, M. J., Feres, M., Figueiredo, L. C.,
Duarte, P. M., Lira, E. A. & Faveri, M. (2010)
Short-term benefits of the adjunctive use of
metronidazole plus amoxicillin in the microbial
profile and in the clinical parameters of
subjects with generalized aggressive periodonti-
tis. Journal of Clinical Periodontology 37, 353–
365.
Nascimento, G. G., Leite, F. R., Do, L. G.,
Peres, K. G., Correa, M. B., Demarco, F. F. &
Peres, M. A. (2015) Is weight gain associated
with the incidence of periodontitis? A system-
atic review and meta-analysis. Journal of Clini-
cal Periodontology 42, 495–505.
Perri, R., Nares, S., Zhang, S., Barros, S. P. &
Offenbacher, S. (2012) MicroRNA modulation
in obesity and periodontitis. Journal of Dental
Research 91, 33–38.
Pradeep, A. R., Kumari, M., Kalra, N. &
Priyanka, N. (2013) Correlation of MCP-4 and

© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
1012 Maciel et al.
high-sensitivity C-reactive protein as a marker
of inflammation in obesity and chronic peri-
odontitis. Cytokine 61, 772–777.
Sanz, Y. & Moya-P!erez, A. (2014) Microbiota,
inflammation and obesity. Advances in
Experimental Medicine and Biology 817, 291–
317.
Socransky, S. S., Haffajee, A. D., Cugini, M. A.,
Smith, C. & Kent, R. L., Jr (1998) Microbial
complexes in subgingival plaque. Journal of
Clinical Periodontology 25, 134–144.
Socransky, S. S., Smith, C., Martin, L., Paster, B.
J., Dewhirst, F. E. & Levin, A. E. (1994)
“Checkerboard” DNA-DNA hybridization.
BioTechniques 17, 788–792.
Suvan, J., D’Aiuto, F., Moles, D. R., Petrie, A. &
Donos, N. (2011) Association between
Clinical Relevance
Scientific rationale for the study:
Evidence suggests a relationship
between periodontal diseases and
obesity. Nevertheless, the impact of
obesity on the subgingival micro-
bial composition has received little
attention.
overweight/obesity and periodontitis in adults.
A systematic review. Obesity Reviews 12, e381–
e404.
Tagliabue, C., Principi, N., Giavoli, C. & Espos-
ito, S. (2016) Obesity: impact of infections and
response to vaccines. European Journal of Clini-
cal Microbiology & Infectious Diseases 35, 325–
331.
World Health Organization (2008) Waist Circum-
ference and Waist–Hip Ratio: Report of a
WHO Expert Consultation: Geneva, December:
1-47.
Zimmermann, G. S., Bastos, M. F.,
Dias Gonc!alves, T. E., Chambrone, L. &
Duarte, P. M. (2013) Local and circulating
levels of adipocytokines in obese and normal
Principal findings: Obesity affects the
subgingival microbial composition,
especially in patients with CP, as this
systemic condition is associated with
increased levels and proportions of
periodontal pathogens.
Practical implications: The negative
impact of obesity on the
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
weight individuals with chronic periodontitis.
Journal of Periodontology 84, 624–633.
Address:
Poliana Mendes Duarte
Universidade Guarulhos, Centro de P! os-
Graduac!a~o e Pesquisa
Prac!a Teresa Cristina, 229 – Centro,
Guarulhos
CEP: 07.023-070, SP
Brazil
E-mail: pduarte@ung.br
composition of subgingival micro-
biota might directly interfere with
periodontal breakdown as well as
with preventive and treatment
strategies for patients presenting
this condition.