You are on page 1of 8

Volume 86 • Number 9

Adiposity Measurements and Non-


Surgical Periodontal Therapy Outcomes
Wiem Bouaziz,* Jean-Luc Davideau,*† Henri Tenenbaum,*† and Olivier Huck*†

Background: Obesity is considered a risk factor for peri-


odontitis. However, its influence on periodontal therapy has
not been clearly determined. The aim of this case-control
study is to evaluate the association between adiposity mea-
surements, non-surgical periodontal treatment outcomes,
and influencing factors in patients with chronic periodontitis.
Methods: Eighteen obese and 18 normal-weight (NW) pa-

R
ecent systematic reviews and meta-
tients are included in this study. The waist/hip ratio (WHR), analyses demonstrated a link among
plaque index, bleeding on probing, probing depth (PD), overweight, obesity, and periodo-
and clinical attachment level (CAL) were measured at base- ntal diseases.1,2 Obesity has been recog-
line and 3 and 6 months after treatment. Univariable and nized as a major public health problem.3
multivariable analyses were used to evaluate the influence Obesity is defined as an abnormal accu-
of sex, age, baseline percentage of PD >3 mm, WHR, and mulation of body fat and is associated with
obesity on periodontal treatment outcomes. increased risk of illness, disability, and
Results: Demographic and periodontal characteristics at death and is considered a risk factor for
baseline were similar in both groups. All periodontal param- cardiovascular diseases, some cancers,
eters were improved during treatment in both groups. PD re- and type 2 diabetes.4,5 People are con-
duction and CAL gain were 0.88 and 0.84 mm in NW sidered overweight (preobese) if their
individuals and 0.79 and 0.68 mm in obese individuals. body mass index (BMI) stands between
The difference in moderate-to-deep pocket (PD >5 mm) per- 25 and 30 kg/m2 and obese when BMI is
centages between the baseline and 6-month examinations >30 kg/m2. However, there are some
was 9.1% in NW individuals and 6.08% for obese individuals. limitations to the use of BMI. In fact, BMI
Multivariable analysis showed that obesity negatively influ- does not measure the corpulence of the
enced changes of PD >5 mm percentages. This influence individual and provides only limited data
was also observed at 3 months for improving sites (PD de- on body composition that varies with
crease >2 mm between examinations) if WHR was also con- age, sex, and physical activity.6 Recent
sidered in the analysis. data suggest that waist circumference
Conclusions: A negative association between adiposity and waist/hip ratio (WHR) are more
measurements and periodontal treatment outcomes was ob- convenient indicators of abdominal fat
served mainly for moderate-to-deep pockets. Consideration distribution and its associated health
of WHR and other influencing factors amplified the negative problems.5
effect of obesity on periodontal treatment outcomes. J Peri- Periodontitis is an infectious and in-
odontol 2015;86:1030-1037. flammatory disease that is one of the
most common chronic diseases through-
KEY WORDS
out the world7 and is closely related to
Obesity; periodontal diseases; treatment outcome; waist-hip diabetes8 and stroke.9 These diseases
ratio. have some common risk factors, such as
smoking.10 Periodontitis appears to affect
* Department of Periodontology, Dental Faculty, Strasbourg, France. more frequently and be more severe in
† National Institutes of Health and Medical Research, Osteoarticular and Dental
Regenerative Nanomedicine, Strasbourg, France. obese individuals than in normal-weight
(NW) individuals.1 The risk factor het-
erogeneity of the observed population in
the different investigations on obesity and
adiposity effects could affect the magni-
tude of the increased risk rather than

doi: 10.1902/jop.2015.140734

1030
J Periodontol • September 2015 Bouaziz, Davideau, Tenenbaum, Huck

precluding the risk.11 Some systemic conditions as- level (CAL) ‡3 mm19; 2) ‡12 teeth; 3) aged >18
sociated with adiposity, such as metabolic syndrome, years; 4) non-smoker (or former smoker since ‡5
dyslipidemia, and insulin resistance, are able to influ- years); 5) have a regular medical follow-up (at least
ence this relationship.12 Interestingly, increased WHR one time per year); and 6) have recent blood samples
seemed to exacerbate the association between obesity that enable control of glycemic or hemoglobin A1c
and periodontal disease.13,14 (HbA1c) (in the previous 3 months).
Presently, few studies have been conducted on the Patients were excluded from the study for the
effect of obesity on periodontal treatment. Some following: 1) diagnosed with diabetes mellitus and
studies have shown that obesity negatively influ- HbA1c level >6.5%; 2) chronic inflammatory disease
enced periodontal treatment outcomes,15,16 but or infection; 3) pregnancy or lactation; 4) physical or
other studies did not demonstrate such an influ- mental handicaps that could interfere with adequate
ence.17,18 The inclusion of other risk factors in some oral hygiene control; 5) currently smoking or a former
studies, such as diabetes18and smoking,15,18 may smoker of <5 years; 6) the use of systemic or topical
influence the results. Furthermore, the variability of steroidal and non-steroidal anti-inflammatory drugs
demographic characteristics, such as age and sex, or antibiotics during the past 6 months; 7) any imm-
and the definition of periodontal status, treatment unologic condition requiring prophylactic antibiotic
modalities (2-15 to 6-month16 follow-up), and out- treatment before invasive dental procedures; 8) any
comes could also modify data interpretation. Finally, history of periodontal treatment; and 9) any diet ful-
the definition of obesity and control patients varied filled to lose weight (or intention to fulfill a diet).
between the different studies, with some basing the Periodontal measurements and anthropometric data
distinction on BMI alone15,16 and others using BMI in were collected by the same trained examiner (WB). The
association with other adiposity measurements data collection was done at baseline and 3 and 6
(WHR, body fat),17,18 which could explain in part the months. At baseline, non-surgical treatment was per-
variability of results1 and the need for additional formed, including oral hygiene education and scaling
studies considering the influence of these factors. and root planing (SRP). First, a trained periodontist
The aim of this study is to evaluate the association (WB) instructed the patient regarding brushing tech-
among adiposity measurements, non-surgical peri- nique (Bass technique) and use of interproximal hy-
odontal treatment outcomes, and influencing factors giene devices. SRP was performed using an ultrasonic
in patients suffering from moderate-to-severe chronic device‡ and manual curets.§ Treatment was performed
periodontitis (CP). in two to three sessions under local anesthesia for
probing depth (PD) >5 mm lasting 1 hour each. The
MATERIALS AND METHODS aim of each session was to remove biofilms and cal-
Study Groups culus of scaled roots. All sessions were performed
Patients diagnosed at the Department of Periodon- within 7 days. Patients were instructed to rinse with
tology, Dental Faculty, University of Strasbourg, a chlorhexidine mouthwash (0.12%)i twice a day for 7
France, as having moderate-to-severe generalized days. The treatment protocol did not include antibiotics
CP are included in this study. The study protocol was treatment. Oral hygiene was controlled at each ap-
accepted by the Ethical Committee of the Faculty of pointment, and hygiene instructions were repeated if
Medicine (2014-61), University of Strasbourg. This needed. At the 3-month examination, reinstruction of
study was conducted from January 2011 to June oral hygiene methods was performed if needed, and
2012. Participants were informed about the objec- residual pockets ‡4 mm were systematically scaled
tives of the study and its design and provided written and planed.20 All treatments procedures were per-
informed consent. The patients (20 males and 20 formed by the same periodontist (WB).
females, aged 22 to 78 years; mean age: 51 – 15
Data Collection
years) were selected and placed in two groups: 1) an
Periodontal examination. To calibrate the examiner,
obese group including patients with BMI ‡30 and 2)
a total of 250 sites were evaluated randomly in six
an NW group including patients with a BMI ‡18.5 and
patients (one quadrant per patient). The examiner
£25. Obese patients were first enrolled and matched
(WB) evaluated the PD parameters on two occasions
by sex and age with NW patients. The maximum
at 1-week intervals with an agreement >0.90 as
allowable difference between an obese patient and
calculated with the k-light score.
the corresponding matched NW patient was 10 years.
Full-mouth clinical measurements included the
The matching ratio was 1:1.
following: 1) plaque index (PI); 2) bleeding on probing
To be enrolled in the study, participants from
both groups had to fulfill all the inclusion criteria: 1)
‡ Suprasson Newtron, Satelec, Acteon, Merignac, France.
a clinical diagnosis of moderate-to-severe general- § Hu-Friedy, Chicago, IL.
ized CP with ‡30% of sites with clinical attachment i Eludril, Pierre Fabre, Castres, France.

1031
Adiposity and Midterm Periodontal Treatment Outcomes Volume 86 • Number 9

dicators of overall adiposity. The


weight was obtained using a cali-
brated scale. The height was
measured using a measuring
board. BMI was calculated with
the computed formula as weight
in kilograms divided by the
square height in meters. Waist
measurements were taken at the
level of the umbilicus in centi-
meters, and hip measurements
were taken at the greatest cir-
cumference with a measuring
tape. Then, the waist was divided
by the hip measurements to ob-
tain the WHR.6
Statistical Analyses
The sample size for intergroup
comparisons was calculated con-
sidering a mean PD reduction of
1 mm from baseline to 6 months
Figure 1.
Flowchart of the study.
with a standard deviation of 0.5
mm21 and with a difference >0.3
mm between groups.18 The sam-
(BOP); 3) PD (in millimeters); and 4) CAL (in milli- ple size was estimated to be 18 patients per group with
meters). PI, BOP, PD, and CAL measurements were a statistical power of 75% and an a error of 0.05.#
performed using a manual periodontal probe¶ at six Considering that 10% of the included patients could
sites per tooth. PD was defined as the distance from drop out during the study period, 20 patients were
the free gingival margin to the bottom of the pocket/ included in each group. A per-protocol analysis has
sulcus, and CAL was defined as the distance from the been performed, and only data of patients that
cemento-enamel junction to the bottom of the pocket/ followed all the study protocol were considered for
sulcus.17 analysis.
Anthropometric measurements. Weight (kilo- Selected periodontal outcomes, including the
grams), height (meters), and waist and hip circum- change of mean PD, mean percentages of PD >5 mm,
ferences (centimeters) were measured at each and the number of improving sites defined as a PD
appointment. BMI and WHR were assessed as in- decrease >2 mm between the baseline and 3- and 6-
month examinations, were analyzed for each patient.
Statistical analysis was performed at the patient level.
Comparison between the NW and obese groups were
Table 1.
performed using Wilcoxon signed-rank test (matched
Demographic Data at Baseline pairs). Comparisons between the baseline and 3-
and 6-month examinations were performed using
Baseline Friedman test with multiple pairwise comparisons
and Bonferroni correction.
Variable NW (n = 18) Obese (n = 18)
Explanatory variables included obesity, age, sex,
Age (years) 51.00 – 13.1 51.3 – 16.7 WHR, and percentage of PD >3 mm at baseline. The
primary variable of interest was obesity, with other
Females (n) 10 8
variables that could influence obesity morbidity
Weight (kg) 66.4 – 12.0 91.7 – 10.9* (WHR, sex) and periodontal treatment outcomes
(initial extent of periodontal disease represented here
Height (m) 1.67 – 0.10 1.66 – 0.09
by the percentage of PD >3 mm, age, and sex).
BMI 23.6 – 1.92 33.3 – 4.03* Random-component mixed models were used for
univariable and multivariable analyses to evaluate
WHR 0.85 – 0.08 0.95 – 0.08*
Data are presented as mean – SD unless otherwise noted. ¶ PCPN115BR, Hu-Friedy.
* P <0.05, significant difference between the NW and obese groups. # G*Power v.3.1.9.2, University of Düsseldorf, Düsseldorf, Germany.

1032
J Periodontol • September 2015 Bouaziz, Davideau, Tenenbaum, Huck

the influence of these factors. Explanatory variables

Diff 6M NW Diff 6M Obese

37.66 – 21.53 33.98 – 17.97

24.14 – 20.72 22.31 – 10.38


0.27 – 0.57
0.95 – 0.62

0.79 – 0.60

6.08 – 8.87

0.68 – 0.74
achieving statistical significance at the 0.20 level in
the univariable analysis for at least one of the se-
lected variables were included in multivariable
analysis. WHR was forced in the different models
considering its effect on obesity morbidity.13 Differ-
0.27 – 0.75
0.76 – 0.39

0.88 – 0.98

9.10 – 14.3

0.84 – 0.78
ences were considered as significant when P <0.05.
6 Months, Mean – SD

All of the analyses were performed using statistical


software.**

† RESULTS
Obese (n = 18)

15.71 – 13.08*
0.41 – 0.45*†
10.36 – 9.92*

2.43 – 0.49*

1.30 – 2.21*

3.37 – 1.15*
Patients’ Demographic Characteristics
23.16 – 5.79

During the study, two patients dropped out in each


group for the following reasons: 1) two patients mi-
grated; 2) one missed the 3-month examination; and
3) one missed the final examination. The flowchart of
0.44 – 0.43*†
NW (n = 18) Obese (n = 18) NW (n = 18) Obese (n = 18) Diff 3M NW Diff 3M Obese NW (n = 18)

7.00 – 7.26*

2.45 – 0.53*
12.76 – 11.0*
2.81 – 3.56*

3.23 – 0.98*

the study is presented in Figure 1. The data of the


25.05 – 2.53

remaining 18 patients in each group were used for the


statistical analyses (Table 1). At the initial exami-
nation, demographic data were statistically compa-
rable in both groups (P >0.05). The mean age was
30.35 – 20.35 29.15 – 18.68

51.3 years for obese individuals and 51 years for NW


0.22 – 0.55
0.53 – 0.49

0.44 – 0.58
14.6 – 8.08
4.81 – 7.83

0.35 – 0.48

individuals. The percentage of males was 56% for


obese versus 44% for non-obese. The weight, BMI,
and WHR were significantly (P <0.01) higher in the
Diff 3M = difference between baseline and 3-month data; Diff 6M = difference between baseline and 6-month data.
Clinical Parameters at Baseline and 3 and 6 Months and Changes

obese group than in the NW group. WHR was sig-


36.89 – 20.85 38.02 – 16.43 17.80 – 10.4* 23.36 – 14.75* 19.09 – 17.46
8.08 – 13.24
0.22 – 0.73
0.45 – 0.30

0.61 – 0.94

0.53 – 0.70

nificantly higher in males (0.95) than in females


3 Months, Mean – SD

(0.86) in the whole group. In the obese group, the


mean BMI was 33.3 (30.12 to 45.35), and the WHR
was 0.95 (0.97 for males, 0.92 for females). During
the study period, patients’ weight significantly in-
0.84 – 0.49*
44.66 – 23.51 44.33 – 19.80 14.31 – 8.09* 15.18 – 9.88*

2.58 – 3.06*

3.70 – 1.14*

creased (<1 kg) in the obese group, but WHR did not
23.22 – 5.75

2.78 – 0.55

change.
Periodontal Characteristics at Baseline and
Response to Non-Surgical Periodontal Treatment
0.75 – 0.54*

3.54 – 0.86*

Periodontal characteristics at baseline were similar


25.11 – 2.54

2.71 – 0.40

3.82 – 3.66

(P >0.05) in both groups for the mean number of teeth


per patient, the mean PI, and the BOP percentage
(Table 2). Mean PD of 3.33 and 3.22 mm and mean
CAL of 4.07 and 4.04 mm were observed in the NW
and obese groups, respectively. At the final exami-
23.44 – 5.53
1.37 – 0.52

3.22 – 0.76

7.39 – 9.78

4.04 – 1.17

nation, all periodontal parameters were improved in


Baseline, Mean – SD

both groups. PI continuously decreased during the


† P <0.05, significant difference from 3 months.
* P <0.05, significant difference from baseline.

study period, but the improvement of other param-


eters was mainly observed at 3 months (BOP, PD >3
11.90 – 14.58

mm percentages, mean CAL) or between 3 and 6


25.33 – 2.63
1.20 – 0.70

3.33 – 0.92

4.07 – 1.15

months (mean PD, PD >5 mm percentage). The re-


duction of mean PD and CAL gain were 0.88 and
0.84 mm for the NW group and 0.79 and 0.68 mm for
the obese group, but these differences between
Mean CAL (mm)

groups were not significant. The differences of PD >3


Mean PD (mm)
PD >3 mm (%)
PD >5 mm (%)

mm and PD >5 percentages between baseline and the


Parameters

Teeth (n)

BOP (%)
Table 2.

final examination were 24.14% and 9.1% for the NW


group and 22.31% and 6.08% for the obese group.
PI

** XLSTAT, Addinsoft, Paris, France.

1033
Adiposity and Midterm Periodontal Treatment Outcomes Volume 86 • Number 9

The numbers of improving sites at the final ex-

Bold type indicates statistically significant differences at P £0.2, and, other than for % PD >3 mm, indicate the parameters that will be included in multivariate analysis models. Negative values indicat poorer
<0.001
0.14
0.40

0.57

0.25
amination were 18.0 and 10.5 in the NW and

P
obese groups, respectively.

20.515 (-51.350 to 92.380)


-9.701 (-22.703 to 3.300)

-7.378 (-20.087 to 5.332)


72.130 (43.770 to 100.49)
0.192 (-0.269 to 0.653)
Nb DPD >2 6M Factors Influencing Periodontal Parameter
Changes: Univariable and Multivariable
Analyses
Univariable analysis showed that the percentage of PD
>3 mm at baseline markedly influenced periodontal
outcomes (Table 3). All periodontal parameters
0.001

changed more favorably in patients suffering from


0.19

0.16
0.83

0.54
P

more severe periodontitis at baseline. Other patient


-8.056 (-20.329 to 4.218)

20.437 (-46.665 to 87.54)

-8.389 (-20.184 to 3.406)


52.404 (22.633 to 82.174)

characteristics, such as sex, age, obesity, and WHR,


0.044 (-0.370 to 0.457)
Univariable Analysis of Periodontal Parameter Changes During Treatment [value (95% CI)]
Nb DPD >2 3M

were not associated with periodontal parameter


changes using univariable analysis.
However, considering the potential influencing
factors, multivariable analysis showed that lower PD
>5 mm percentage changes were significantly ob-
M = months; DPD = reduction of mean PD; D%PD >5 = percentage reduction of PD >5 mm; Nb DPD = number of improving PD >2 mm sites.

served in females and in obese patients at 3 and 6


<0.001
0.15
0.65

0.74

0.24

months. The negative effect of obesity on PD >5 mm


P

percentages changes and improving site numbers


0.001 (-0.003 to 0.004)

-0.070 (-0.167 to 0.027)

0.055 (-0.284 to 0.395)

-0.028 (-0.075 to 0.019)


0.396 (0.274 to 0.519)

was amplified if WHR was included in multivariable


D%PD >5 6M

analysis. Obesity appeared to be an independent


predictor of poorer periodontal treatment outcomes
at midterm (Table 4).

DISCUSSION
<0.001
0.14
0.68

0.87

0.23

The present study evaluates the influence of obesity


P

on the response to non-surgical treatment in pa-


-0.064 (-0.149 to 0.021)

0.026 (-0.297 to 0.349)

-0.028 (-0.074 to 0.018)

tients with CP. Both conditions are chronic in-


0.382 (0.264 to 0.499)
0.001 (-002 to 0.004)

flammatory diseases that can potentially burden the


D%PD >5 3M

systemic inflammatory response.11,22 The results


showed an association between adiposity mea-
surements and non-surgical periodontal treatment
outcomes at 3 and 6 months in patients suffering
from moderate-to-severe CP. This study confirmed
<0.001
0.14

0.16

0.27

0.71
P

that non-surgical periodontal therapy promoted


a significant improvement in all periodontal clinical
0.014 (-0.005 to 0.034)

-0.413 (-0.994 to 0.169)

1.609 (-1.290 to 4.509)

-0.092 (-0.587 to 0.403)


2.770 (1.615 to 3.925)

parameters, such as PI, BOP, PD, and CAL, at 6


months, regardless of the adiposity status, as ob-
DPD 6M

served previously in comparable studies.16-18 These


overall good results are partially attributable to
positive patient cooperation. Obese and NW pa-
tients were equally receptive to hygiene instructions
<0.001
0.11

0.21
0.25

0.51

and demonstrated motivated behavior, as shown


P

here by the continuous PI reduction during treat-


0.010 (-0.008 to 0.029)

-0.422 (-0.945 to 0.101)

1.830 (-1.073 to 4.733)

-0.171 (-0.696 to 0.354)

ment. However, comparison between groups


2.612 (1.45 to 3.773)

showed that the improvement of other periodontal


DPD 3M

treatment outcomes, such as mean PD, percentage


of PD >5 mm changes, and number of improving
sites, was less pronounced in the obese group than
clinical responses.

in the NW group. However, these differences be-


tween groups were under the level of significance.
% PD >3 mm
at baseline
Sex (female)
Table 3.

Covariable

Previous studies demonstrated a differential treat-


Obese
WHR

ment response in obese patients,15,16 whereas other


Age

studies showed the contrary.17,18 Most of the studies

1034
J Periodontol • September 2015 Bouaziz, Davideau, Tenenbaum, Huck

were performed short term (£3 months)15,17,18 or for

0.001

<0.001
6 months.16 In the present investigation, various

0.98
0.25

0.07

0.76
0.48

0.44
0.06
P
kinetics of periodontal parameter changes during

31.970 (-50.507 to 114.44)


the 6-month examination period were observed. In

-9.644 (-20.093 to 0.805)

-4.476 (-17.139 to 8.188)

-11.961 (-24.451 to 0.529)


50.59 (21.436 to 79.747)

69.35 (41.015 to 97.687)


-0.005 (-0.369 to 0.358)
-6.174 (-16.89 to 4.551)

0.060 (-0.330 to 0.450)


the study by Goncxalves et al.,16 the significant effect
Nb DPD >2 6M

of obesity was only observed at 6 months and not


at 3 months, suggesting that intervals between
the baseline and final examinations may influence
results.
This study showed that obesity was significantly
<0.001

<0.001

0.02
associated with PD >5 mm percentage changes and
0.48
0.17

0.08

0.51
0.85

0.15
P

numbers of improving sites. In previous studies,

-14.756 (-26.730 to -2.781)


-1.116 (-13.257 to 11.024)

57.58 (-21.486 to 136.65)


Multivariable Analysis of Periodontal Parameter Changes During Treatment [value (95% CI)]

significant associations with obesity and/or BMI


-7.284 (-17.729 to 3.162)

-9.123 (-19.300 to 1.053)


68.43 (40.038 to 96.830)

52.24 (25.076 to 79.409)


0.125 (-0.229 to 0.479)

-0.122 (-0.496 to 0.252)

were observed for other periodontal treatment out-


Nb DPD >2 3M

comes, such as mean PD and/or percentages of


residual pockets at the final examination.15,16 In
these studies, the strength of the association with
obesity and BMI varied depending on periodontal
M = months; DPD = reduction of mean PD; D%PD >5 = percentage reduction of PD >5 mm; Nb DPD = number of improving PD >2 mm sites.

treatment outcomes and was higher for percent-


<0.001

<0.001

0.005
0.04

0.01

ages of PD >4 mm than for the mean PD15 or in-


0.59

0.78
0.16

0.18
P

versely.16 Different thresholds of PD have been used


-0.062 (-0.122 to -0.002)

-0.040 (-0.070 to -0.010)

-0.056 (-0.094 to -0.018)

in previous studies: PD £3 mm,18 PD >4 mm,15-17 and


0.001 (-0.001 to 0.002)

0.00 (-0.002 to 0.002)


-0.045 (-0.109 to 0.019)

0.190 (-0.092 to 0.472)


0.385 (0.276 to 0.495)

0.388 (0.280 to 0.496)

PD >6 mm.16 In the present study, a PD >5 mm


D %PD >5 6M

threshold was chosen, although the persistence of PD


>5 mm after initial treatment has been associated
Bold type indicates statistically significant differences at P £0.05. Negative values indicat poorer clinical responses.

with disease progression and increased tooth loss at


long term.23 Furthermore, the present study indicates
<0.001

<0.001

that the number of improving sites decreased with


0.03

0.04

0.02
0.51

0.70
0.11

0.08
P

obesity. Improving site evaluation could readily de-


tect significant clinical changes.24 Indeed, in the
0.000 (-0 0.001 to 0.002)
-0.056 (-0.106 to -0.007)

-0.041 (-0.073 to -0.008)

-0.052 (-0.094 to -0.011)


0.001 (-0.001 to 0.002)

0.135 (-0.160 to 0.025)

study by Preshaw et al.,24 the difference of improving


0.378 (0.270 to 0.486)

0.383 (0.274 to 0.491)


-0.045 (-0.1 to 0.011)
D %PD >5 3M

site showed the effect of quitting smoking during


periodontal treatment, whereas mean PD changes
did not. Interestingly, in the present study, mean PD
is not associated with obesity. Previous data have
shown that mean PD changes did not demonstrate
<0.001

<0.001
0.256
0.12
0.13

0.39

0.41

0.41
0.24

the strongest association with various explanatory


P

variables, such as obesity15 and interleukin-6 (IL-6)


0.012 (-0.003 to 0.027)
-0.345 (-0.797 to 0.108)

-0.165 (-0.550 to 0.221)

0.009 (-0.007 to 0.026)


-0.224 (-0.764 to 0.317)

1.371 (-1.928 to 4.669)


-0.285 (-0.768 to 0.198)

polymorphism.21 The choice of periodontal param-


2.565 (1.411 to 3.719)

2.600 (1.438 to 3.762)

eters to evaluate treatment outcomes could appar-


D PD 6M

ently influence the type of relationship with obesity. It


could also not be excluded that the number and the
profile of patients included in the present study does
not allow the determination of an effect on mean PD
<0.001

<0.001
0.27
0.12

0.24

0.63
0.56

0.20
0.09

changes.
P

In studies about the effect of obesity on periodontal


treatment outcomes,15-18 modifying/risk factors for
0.008 (-0.007 to 0.023)
-0.343 (-0.094 to 0.780)

-0.239 (-0.646 to 0.168)

-0.004 (-0.012 to 0.020)


-0.152 (-0.676 to 0.372)

2.162 (-1.163 to 5.488)


-0.430 (-0.927 to 0.066)
2.399 (1.235 to 3.563)

2.465 (1.311 to 3.619)

periodontal treatment outcomes and for adiposity


D PD 3M

morbidity may also influence the results. Regarding


both the well-established negative effect of diabetes
on periodontal treatment25 and its high prevalence in
obese patients,26,27 patients with diabetes have been
% PD >3 mm
%PD >3 mm

excluded in the present study, as has been done in


at baseline

at baseline
Sex (female)

Sex (female)
Table 4.

some other similar investigations.15-17 To eliminate


Obese

Obese
Model 1

Model 2

WHR
Age

Age
Model

potential undiagnosed diabetes, HbA1c level has


been assessed before the investigation, and patients

1035
Adiposity and Midterm Periodontal Treatment Outcomes Volume 86 • Number 9

with diabetes (HbA1c >6.5%) were excluded, as was decreases serum levels of tumor necrosis factor-a,
done in the study by Gonc xalves et al.16 Smokers have IL-6, and leptin.18 Interestingly, periodontal treat-
been also excluded. 16,17 Univariable analysis of the ment did not decrease others inflammatory markers,
studied population showed that the initial severity of such as C-reactive protein (CRP), or did not modify
periodontal disease (percentage of PD >3 mm) posi- the lipid profile at 3 months.18 This observation may
tively influenced changes of deep pockets and the be explained by the continuous production of in-
number of improving sites. Obesity alone did not in- flammatory products by adipocytes30 that may affect
fluence periodontal outcomes, as shown previously.15 periodontal tissue response as suggested for patients
In multivariable analysis models taking into account with diabetes.31 Interestingly, the nullification of systemic
age, sex, and initial severity of periodontitis, a signifi- effects by obesity in response to periodontal treatment
cant negative influence of obesity was observed for has been demonstrated in the PAVE (Periodontitis
deep pocket changes, but this influence was not ob- and Vascular Events) study for high-sensitivity CRP
served for mean PD and percentages of PD >3 mm reduction.32
changes (data not shown). Similarly, in the study by Some limitations could be attributed to the present
Gonc xalves et al.,16 the observed difference of mean PD study. The specific profile of the obese patients (non-
reduction between obese and non-obese patients is diabetic, non-smoker) may restrain the extrapolation
more marked for initially deep sites than for initially of the results to the whole obese population. Fur-
moderate sites.16 This could be explained by the se- thermore, the small sample size and the impossibility
verity of the initial periodontal status, although initial to achieve investigator masking regarding the visible
mean PD and percentage of PD >4 mm appeared more aspect of obesity may also limit the strength of this
severe in the study by Suvan et al.15 (4 mm and 50%), study. However, the matched design (for age and
in the study by Gonc xalves et al.16 (3.5 mm and 23.5%), sex), the absence of other risk factors (smoking,
and in the study by Altay et al.18 (3.1 mm and 18%) diabetes) in the studied population, and the stan-
compared with the present study (3.2 mm and 17.5%) dardization of periodontal treatment modalities may
(data not shown). These data suggested that the effect reduce the influence of those issues.
of obesity was more pronounced in the case of severe
periodontitis. CONCLUSIONS
In previous studies, the definition of obese versus Within the limitations of this study, it could be con-
non-obese patients was only based on BMI, as in the cluded that adiposity measurements and non-surgical
study by Suvan et al.,15 or did not include other periodontal treatment outcomes were associated
adiposity measurements, such as body fat,18 WHR,16 in patients suffering from moderate-to-severe CP.
or both.17 Interestingly, the inclusion of the WHR Obese patients without diabetes had a poorer response
variable in multivariable analysis amplified the neg- to initial periodontal treatment than NW patients. This
ative association between periodontal treatment effect was mainly observed for moderate-to-deep
outcomes and obesity. Previous studies have shown pockets, suggesting that, the more severe peri-
that WHR or waist circumference could independently odontitis was, the more negative the effect of adiposity
influence periodontal status. 13,14 However, in the was on periodontal treatment outcomes. Interestingly,
present study, WHR alone is not associated with the consideration of WHR and other influencing factors
periodontal treatment outcomes. The study by Jimenez amplified the negative effect of obesity on periodontal
et al.13 has shown that WHR exacerbated the link treatment outcomes.
between periodontitis and obesity in males. These
results suggested that the definition of obesity and ACKNOWLEDGMENTS
adiposity could influence their predictive value for The authors thank Dr. Francxois Séverac (Department
periodontal treatment outcomes. Adiposity contrib- of Public Health, Hôpitaux Universitaires de Stras-
uted to systemic inflammation through modulation of bourg, Strasbourg, France) for his helpful comments.
immune and metabolic parameters, increasing peri- The authors report no conflicts of interest related to
odontitis susceptibility,18 including resistin, leptin, and this study.
adiponectin.28 It has been demonstrated previously
that periodontal treatment outcomes may be influ- REFERENCES
enced by systemic inflammation, and, conversely, 1. Chaffee BW, Weston SJ. Association between chronic
periodontal treatment has systemic consequences as periodontal disease and obesity: A systematic re-
demonstrated in systemically compromised patients, view and meta-analysis. J Periodontol 2010;81:
especially patients with diabetes.29 In obese patients, 1708-1724.
2. Keller A, Rohde JF, Raymond K, Heitmann BL. The
several studies have been conducted to assess association between periodontal disease and over-
the systemic effects of periodontal treatment. It was weight and obesity: A systematic review. J Periodontol
demonstrated that periodontal treatment significantly 2015;12:1-15.

1036
J Periodontol • September 2015 Bouaziz, Davideau, Tenenbaum, Huck

3. World Health Organization. Obesity: Preventing and 19. Armitage GC. Development of a classification system
managing the global epidemic. Report of a WHO for periodontal diseases and conditions. Ann Periodontol
consultation. World Health Organ Tech Rep Ser 2000; 1999;4:1-6.
894: 1-xii, 1-253. 20. Leininger M, Tenenbaum H, Davideau JL. Modified
4. Seo MH, Rhee EJ. Metabolic and cardiovascular impli- periodontal risk assessment score: Long-term predic-
cations of a metabolically healthy obesity phenotype. tive value of treatment outcomes. A retrospective
Endocrinol Metab (Seoul) 2014;29:427-434. study. J Clin Periodontol 2010;37:427-435.
5. Song X, Jousilahti P, Stehouwer CDA; DECODE Study 21. D’Aiuto F, Ready D, Parkar M, Tonetti MS. Relative
Group. Cardiovascular and all-cause mortality in re- contribution of patient-, tooth-, and site-associated vari-
lation to various anthropometric measures of obesity in ability on the clinical outcomes of subgingival debride-
Europeans. Nutr Metab Cardiovasc Dis 2015;25: ment. I. Probing depths. J Periodontol 2005;76:398-405.
295-304. 22. Falagas ME, Kompoti M. Obesity and infection. Lancet
6. World Health Organization. Waist circumference and Infect Dis 2006;6:438-446.
waist–hip ratio: Report of a WHO Expert Consultation. 23. Matuliene G, Pjetursson BE, Salvi GE, et al. Influence of
Geneva: World Health Organization; 2008:1-47. residual pockets on progression of periodontitis and
7. Kassebaum NJ, Bernabé E, Dahiya M, Bhandari B, tooth loss: Results after 11 years of maintenance. J Clin
Murray CJL, Marcenes W. Global burden of severe Periodontol 2008;35:685-695.
periodontitis in 1990-2010: A systematic review and 24. Preshaw PM, Heasman L, Stacey F, Steen N,
meta-regression. J Dent Res 2014;93:1045-1053. McCracken GI, Heasman PA. The effect of quitting
8. Mealey BL, Ocampo GL. Diabetes mellitus and peri- smoking on chronic periodontitis. J Clin Periodontol
odontal disease. Periodontol 2000 2007;44:127-153. 2005;32:869-879.
9. Huck O, Saadi-Thiers K, Tenenbaum H, et al. Evaluat- 25. Costa FO, Miranda Cota LO, Pereira Lages EJ, et al.
ing periodontal risk for patients at risk of or suffering Progression of periodontitis and tooth loss associated
from atherosclerosis: Recent biological hypotheses and with glycemic control in individuals undergoing peri-
therapeutic consequences. Arch Cardiovasc Dis 2011; odontal maintenance therapy: A 5-year follow-up
104:352-358. study. J Periodontol 2013;84:595-605.
10. Hujoel PP, Drangsholt M, Spiekerman C, DeRouen TA. 26. Charles MA. Update on the epidemiology of obesity and
Periodontitis-systemic disease associations in the pres- type 2 diabetes in France (in French). Diabetes Metab
ence of smoking — Causal or coincidental? Periodontol 2000;26(Suppl. 3):17-20.
2000 2002;30:51-60. 27. Kodama S, Horikawa C, Fujihara K, et al. Quantitative
11. Suvan J, D’Aiuto F, Moles DR, Petrie A, Donos N. relationship between body weight gain in adulthood
Association between overweight/obesity and peri- and incident type 2 diabetes: A meta-analysis. Obes
odontitis in adults. A systematic review. Obes Rev Rev 2014;15:202-214.
2011;12:e381-e404. 28. Wozniak SE, Gee LL, Wachtel MS, Frezza EE. Adipose
12. Watanabe K, Cho YD. Periodontal disease and meta- tissue: The new endocrine organ? A review article. Dig
bolic syndrome: A qualitative critical review of their Dis Sci 2009;54:1847-1856.
association. Arch Oral Biol 2014;59:855-870. 29. Corbella S, Francetti L, Taschieri S, De Siena F, Fabbro
13. Jimenez M, Hu FB, Marino M, Li Y, Joshipura KJ. MD. Effect of periodontal treatment on glycemic control
Prospective associations between measures of adipos- of patients with diabetes: A systematic review and meta-
ity and periodontal disease. Obesity (Silver Spring) analysis. J Diabetes Investig 2013;4:502-509.
2012;20:1718-1725. 30. Ortega FJ, Fernández-Real JM. Inflammation in adi-
14. Muñoz-Torres FJ, Jiménez MC, Rivas-Tumanyan S, pose tissue and fatty acid anabolism: When enough is
Joshipura KJ. Associations between measures of enough! Horm Metab Res 2013;45:1009-1019.
central adiposity and periodontitis among older adults. 31. Santacroce L, Carlaio RG, Bottalico L. Does it make
Community Dent Oral Epidemiol 2014;42:170-177. sense that diabetes is reciprocally associated with peri-
15. Suvan J, Petrie A, Moles DR, et al. Body mass index as odontal disease? Endocr Metab Immune Disord Drug
a predictive factor of periodontal therapy outcomes. J Targets 2010;10:57-70.
Dent Res 2014;93:49-54. 32. Offenbacher S, Beck JD, Moss K, et al. Results from the
16. Gonc xalves TE, Feres M, Zimmermann GS, et al. Effects Periodontitis and Vascular Events (PAVE) Study: A
of scaling and root planing on clinical response and pilot multicentered, randomized, controlled trial to
serum levels of adipocytokines in obese patients with study effects of periodontal therapy in a secondary
chronic periodontitis. J Periodontol 2015;86:53-61. prevention model of cardiovascular disease. J Peri-
17. Zuza EP, Barroso EM, Carrareto ALV, et al. The role of odontol 2009;80:190-201.
obesity as a modifying factor in patients undergoing
non-surgical periodontal therapy. J Periodontol 2011; Correspondence: Dr. Olivier Huck, Department of Peri-
82:676-682. odontology, 8 rue Sainte-Elisabeth, F-67000 Strasbourg,
18. Altay U, Gürgan CA, A gbaht K. Changes in inflamma- France. E-mail: huck.olivier@gmail.com.
tory and metabolic parameters after periodontal treat-
ment in patients with and without obesity. J Periodontol Submitted December 21, 2014; accepted for publication
2013;84:13-23. April 3, 2015.

1037