Association Between Periodontitis and Chronic Obstructive Pulmonary Disease in a Chinese Population

Association Between Periodontitisand Chronic Obstructive PulmonaryDisease in a Chinese PopulationYan Si,* Hong Fan,† Yiqing Song,‡ Xuan Zhou,§ Jing Zhang,§ and Zuomin Wang§

Background: A relationship between periodontitis andchronic respiratory disease has been suggested by recentstudies. The aim of this study is to explore the associationbetween periodontitis and chronic obstructive pulmonary dis-ease (COPD) in a Chinese population.

Methods: We conducted a case-control study of 581 COPDcases and 438 non-COPD controls. Lung function examina-tion, a 6-minute walk test, and the British Medical ResearchCouncil questionnaire were performed. Periodontal clinicalexamination index included probing depth (PD), attachmentloss (AL), bleeding index (BI), plaque index (PI), and alveolarbone loss. A validated index for predicting COPD prognosis,the body mass index, airflow obstruction, dyspnea, and exer-cise capacity (BODE) index, was also calculated.

Results: Participants with more severe COPD were morelikely to have severe periodontal disease. PD, AL, PI, alveolarbone loss, and the number of teeth were significantly associ-ated with all stages of COPD (all P <0.001). When comparedto controls (BODE = 0), participants with higher BODE scoreshad significantly higher AL (P <0.001), BI (P = 0.027), PI(P <0.001), alveolar bone loss (P <0.001), and the number ofteeth (P <0.001). PI appeared to be the main periodontalhealth-related factor for COPD, with an odds ratio (OR) =9.01 (95% CI = 3.98 to 20.4) in the entire study populationOR = 8.28 (95% CI = 2.36 to 29.0), OR = 5.89 (95% CI =2.64 to 13.1), and OR = 2.46 (95% CI = 1.47 to 4.10) for cur-rent, smokers, and non-smokers, respectively.

Conclusion: Our study found a strong association betweenperiodontitis and COPD, and PI seemed to be a major peri-odontal factor for predicting COPD among Chinese adults.J Periodontol 2012;83:1288-1296.

KEY WORDS

Periodontitis; pulmonary disease, chronic obstructive;respiratory function tests.

Chronic obstructive pulmonary dis-ease (COPD) is a major globalhealth problem and is a main

cause of death and disability worldwide.1

COPD patients have a compromised lifequality and high mortality rate.2,3 COPDis the fourth leading cause of death in theworld, and its prevalence and mortalityare expected to increase in the comingdecades.4 A COPD patient is character-ized by irreversible airflow limitation,which is usually progressive and causedby an abnormal inflammatory responseof the lung to noxious particles or gases.5

Bacterial infection and smoking are typ-ically considered as main risk factors forCOPD.6

Periodontitis is also a highly prevalenthealth problem and manifests as a chronicinflammatory reaction to bacterial infec-tions. Periodontitis is a destructive dis-ease that affects the tooth-supportingtissues and ultimately leads to tooth lossin 15% of adults.7 The results of the ThirdNational Oral Health Survey of the Minis-try of Health, China, showed that theprevalence of healthy periodontal statuswas only 14.5% for adults 35 to 44 yearsold and 14.1% for those 65 to 74 yearsold.8 Only one seventh of the middle-aged and older adult population had nogingival bleeding, no deep pocket, andno attachment loss (AL) of <3 mm, whichindicated the widespread occurrence ofperiodontitis in China.8

Recent research has suggested thatperiodontitis not only causes gingival

* Department of Preventive Dentistry, Peking University School and Hospital ofStomatology, Beijing, China.

† Department of Stomatology, Shanxi People’s Hospital, Shanxi, China.‡ Division of Preventive Medicine, Brigham and Women’s Hospital, Harvard Medical School,

Boston, MA.§ Department of Stomatology, Beijing ChaoYang Hospital affiliated with Capital Medical

University, Beijing, China.

doi: 10.1902/jop.2012.110472

Volume 83 • Number 10

1288

bleeding but also is a risk factor for systemicinflammation-related chronic diseases, such as car-diovascular disease,9-11 diabetes, and COPD.12-14

Previous studies have also shown an association be-tween poor oral health and COPD, independent ofother confounding variables, such as age, sex, andsmoking.15,16 Patients with COPD were found to havemore mean clinical AL than those without COPD, andthe oral hygiene scores and percentage of gingivalbleeding sites were significantly associated with theseverity of COPD.17 At least one prospective studyfound that the risk of COPD in participants who werecurrent smokers significantly increased with the se-verity of periodontal disease.18 However, the evi-dence has been inconclusive because of limited data.

We conducted a case-control study involving largenumbers of cases and controls with multiple metricsof lung function and COPD prognostic factors ina Chinese population to further investigate the rela-tionship between periodontitis and COPD.

MATERIALS AND METHODS

Study PopulationParticipants of this multicenter study were enrolledfrom outpatients >30 years old (640 males and 379females; mean age: 63.39 – 9.44 years) receivinghealth care at the Departments of Respiratory Medi-cine (patients with COPD; n = 581) and Departmentsof Dentistry (controls; n = 438) in eight hospitals inBeijing, China. Consecutive patients were enrolledfrom March 2007 to December 2009. Patients werenot eligible for the study if they had: 1) < 15 teeth;2) experienced exacerbations of COPD symptoms(including cough or expectoration, pyrexia, or aggra-vation of dyspnea); or 3) changed medication use asa result of respiratory symptoms within 4 weeks be-fore recruitment. Patients who met the eligibility crite-ria, and were willing to participate, signed a consentform. The study protocol was approved by the humanresearch ethical board from Beijing ChaoYang Hos-pital, Beijing, China.

A pretest was performed in a small populationsimilar to the sample of this study to find out if any ques-tions might be unclear, and determine if the structure ofthe questionnaire was acceptable. The useful informa-tion was collected and the draft of the questionnairewas revisedwithcareonwordingandcontent.Theques-tionnaire included demographic variables, such as age,sex, marital status, occupation, educational level, liv-ing condition, smoking status and number of ciga-rettes smoked daily. The British Medical ResearchCouncil (MRC)19 questionnaire was also completed.

Evaluation of Periodontal StatusThe oral examinations were performed by two exam-iners (XZ, JZ), who had received training by an expe-

rienced dentist (ZW). Replicate examinations wereconducted throughout the study to assess intra-examiner reliability. The k value of intra-examineragreement was 0.82. The periodontal examinationincluded probing depth (PD), AL, bleeding index(BI),20 and plaque index (PI),21. BI was scored on a0 to 5 scale, and PI was determined on a 0 to 3 scaleafter air drying. The four indices were recorded on sixsites of each tooth. The highest scores of buccal andlingual surfaces were recorded separately, and thenthe higher of the two was used for analysis. A full-mouth series of intraoral periapical radiographs wereused to assess the level of alveolar bone loss (ABL),with a score of 0 (without ABL), 1 (ABL <1/3 of the rootin length), 2 (ABL from 1/3 to 2/3 of the root in length),or 3 (ABL >2/3 of the root in length) assigned to eachtooth.22

According to the classification of periodontitis usedin the US Third National Health and Nutrition Exami-nation Survey,23 the participants were divided intothree groups: 1) mild periodontitis (n = 232), ‡1 teethwith ‡3 mm PD, <30% of the teeth examined having‡4 mm PD, and no teeth having ‡5 mm PD; 2) mod-erate periodontitis (n = 351), ‡1 teeth with ‡5 mm PDor 30% to 60% of the teeth examined having ‡4 mmPD; or 3) severe periodontitis (n = 436), ‡30% ofthe teeth examined having ‡5 mm PD or ‡60% ofthe teeth examined having ‡4 mm PD.

Test of Lung FunctionLung function of each participant was measured usingspirometry and presented as percentage of forcedexpiratory volume in 1 second (FEV1) over predictedvalue and the ratio of FEV1/forced vital capacity(FEV1/FVC). The control group was in accordancewith the criteria of FEV1/FVC >0.70, and all patientswith COPD were clinically diagnosed and confirmedby lung function examination. Severity of patientswith COPD was classified according to the Global Ini-tiative for Chronic Obstructive Lung Disease (GOLD)definition:6,24,25 1) mild (stage I; FEV1/FVC <0.70,and FEV1 ‡80% predicted ); 2) moderate (stageFEV1/FVC <0.70 and FEV1 from 50% to 80% pre-dicted II; ); 3) severe (stage III; FEV1/FVC <0.70 andFEV1 from 30% to 50% predicted. ); or 4) verysevere (stage I FEV1/FVC <0.70 and FEV1 <30%predicted or FEV1 <50% predicted plus chronic res-piratory failure V; ). Stage I is . Respiratory failure isdefined as arterial partial pressure of oxygen <8.0kPa (60 mmHg) with or without arterial partial pres-sure of CO2 (PaCO2) >6.7 kPa (50 mm/Hg) whilebreathing air at sea level.

All patients underwent a 6-minute walk test(6MWT), which is recognized as a simple and reliablemethod to evaluate lung function and prognosis ofCOPD.26

J Periodontol • October 2012 Si, Fan, Song, Zhou, Zhang, Wang

1289

The MRC questionnaire27 to assess the severity ofbreathlessness was also used. A score ranging from0 (‘‘I only get breathless with strenuous exercise’’)to 4 (‘‘I am too breathless to leave the house or I ambreathless when dressing or undressing’’)27,28 wasrecorded for each patient(score 0:434 [42.6%]pa-tients ; score 1: 303[29.7%]; score 2177 [17.4%];score 3:66 [6.5%]; and score 4: 39 [3.8%]).

Finally, the body mass index (BMI), airflow ob-struction, dyspnea, and exercise capacity (BODE)29

index, a multidimensional grading system that simul-taneously considers airflow obstruction (using FEV1),dyspnea (using MRC score), and exercise capacity(using 6-minute walking distance), a validated tool(BODE)30-32 to help predict COPD prognosis was ap-plied. The criteria for scoring the BODE index isdescribed in Table 1.29 The BODE score of eachparticipant is the sum of thefour variables with scoresranging from 0 to 10, with higher scores indicatingmore severe COPD prognosis.

Participants were divided into four groups: 1)BODE = 0 (n = 429); 2) BODE = 1 or 2 (n = 320);3) BODE = 3 or 4 (n = 180); and 4) BODE ‡5 (n =90). The latter 3 groups could then be combined toone group: BODE >0 (n = 590).

Statistics AnalysesA statistical software packagei was used for data anal-yses. A two-sided P value <0.05 was considered to bestatistically significant. Independent-samples t testwas used to compare continuous variables, and x2 testwas used to compare categorical variables betweenthe COPD and non-COPD groups. One-way analysisof variance (ANOVA) was used to compare the peri-odontal status among different COPD stages and dif-ferent BODE scores. Pearson correlation coefficientsbetween lung function index and periodontal indexwere computed.

Logistic regression analysis was performed to as-sess the odds ratio (OR) and 95% confidence interval(CI) of COPD in association with periodontal healthstatus. The covariates included demographic vari-ables, smoking status, and BMI. BODE score was con-sidered as the dependent variable, participants werealso stratified by smoking status.

RESULTS

Demographic characteristics of the study populationare summarized in Table 2. The mean – SD age of par-ticipants was 63.4 – 9.4 years, and mean – SD BMIwas 25.0 – 4.05. Patients with COPD were more likelyto be men, current or former smokers, or had lower ed-ucation; there were no statistical differences betweenage, BMI, marital status, and living condition betweenthe two groups.

Participants were divided into three groups accord-ing to severity of periodontitis, and patients withCOPD were classified into four groups according toseverity of COPD (Table 3). Overall, participants withmore severe COPD were more likely to have severeperiodontal disease.

The mean PI, AL, ABL, number of teeth, and per-centage of patients with PD ‡4 mm and AL ‡3 mmwere higher in COPD group than in control group,and these indexes increased with severity of COPD.The difference among the different stages of COPDwas statistically significant (P <0.001), which impliedthat the patients with more severe COPD would havehigher PD value and higher percentage of patients withPD ‡5 mm and AL ‡4 mm (Table 4). Patients withhigher BODE scores had significantly higher BI, PI,AL, ABL, and less number of teeth.

The correlation coefficients between periodontalindex and lung function index are presented in Table5. The periodontal indicesPD, PI, AL, and ABL werenegatively correlated with FEV1% predicted andFEV1/FVC, but positively correlated with MRC andBODE scores. Number of teeth was positively corre-lated with FEV1% predicted, FEV1/FVC, and 6MWTbut negatively correlated with MRC and BODE scores.PD was negatively correlated with FEV1% predictedand FEV1/FVC, but not correlated with other lungfunction indices.

Logistic regression analysis showed that PI wasthe main periodontal health-related factor for COPD(Table 6). After adjusting for age, sex, occupation, ed-ucational level, and smoking status, the COPD OR =9.01 (95% CI = 3.98 to 20.4) for PI in the entire sample.When stratified by smoking status, the OR = 8.28 (95%CI = 2.36 to 29.0) among current smokers, OR = 5.89(95% CI = 2.64 to 13.1) among former smokers, andOR = 2.46(95%CI = 1.47 to4.10)amongnon-smokers.Other periodontal indexes also related to COPD, suchas AL (OR = 1.41 in the entire sample, OR = 1.59 in for-mer smokers) and ABL (OR = 3.89 in non-smokers)can be seen in Table 6.

Table 1.

Criteria of Scoring BODE Index

BODE Index

of COPD29 0 1 2 3

FEV1 (% predicted) ‡65 50 to 64 36 to 49 £35

6MWT (m) ‡350 250 to 349 150 to 249 £149

MRC score 0 to 1 2 3 4

BMI (kg/m2) >21 £21 NA NA

NA = not applicable.

i SPSS statistical package, v.12.0, IBM, Chicago, IL.

Periodontitis and Chronic Obstructive Pulmonary Disease Volume 83 • Number 10

1290

DISCUSSION

In this study, a strong associationbetween periodontitis and COPDwas shown. The results showed thatindividuals with higher BODE sco-res had significantly higher BI, PI,AL, ABL, and less number of teeth.This study also showed that PI wasthe main periodontal health-relatedfactor for COPD. These findings in-dicate the importance of promotingoral hygiene in the prevention andtreatment of COPD.

BODE is a multidimensionalgrading system evaluating patientswith COPD that was initially pro-posed in 200429 and has beenwidely used in recent years.30-32

Classification of COPD accordingto the GOLD definition is basedon post-bronchodilator FEV1. Theaggregative indicator (BODE) in-cludesBMI, airflowobstruction,dys-pnea, and exercise capacity. Inaddition to FEV1, BODE also in-cludes other physiologic and clini-cal variables, so it could reflectgeneral characteristics of COPD,and the accuracy of mortality prog-nosis using BODE is better thanusing FEV1 only.29 In this studyof a Chinese population, the resultsfrom BODE are similar to the re-sults of traditional classification ofCOPD stages, so BODE proved tobe a good indicator for evaluatingCOPD, reflecting the aggregativeconditions of patients with COPD,and is also suitable for analyzing

Table 2.

Basic Demographic and Lifestyle Characteristics ofParticipants

Demographic

Variables

Control Group

(n = 438)

COPD Group

(n = 581) P Value

Age (mean – SD) 62.8 – 9.5 63.9 – 9.4 0.06*

Sex <0.001†

Female 220 (50.2%) 159 (27.4%)Male 218 (49.8%) 422 (72,6%)

Marital status 0.10†

Married 391 (89.3%) 537 (92.4%)Unmarried, divorced,

or widowed47 (10.7%) 44 (7.6%)

Occupation <0.001†

Worker 114 (26.0%) 251 (43.2%)Academic faculty/teacher 124 (28.3%) 130 (22.4%)Technician 112 (25.6%) 125 (21.5%)Other occupation 88 (20.1%) 75 (12.9%)

Educational level‡ <0.001†

Primary school or below 49 (11.2%) 113 (19.4%)Middle school 80 (18.3%) 187 (32.2%)Senior high school 119 (27.2%) 114 (24.3%)College or above 190 (43.4%) 140 (24.1%)

Living condition§ 0.09†

Living alone 43 (13.4%) 55 (17.0%)Living with family 55 (17.2%) 167 (51.5%)

Smoking status <0.001†

Current smoker 49 (11.2%) 113 (19.4%)Former smoker 76 (17.4%) 294 (50.6%)Non-smoker 313 (71.5%) 174 (29.9%)

BMI (mean – SD) 25.1 – 3.8 24.9 – 4.2 0.54*

* P value obtained from independent-samples t test.† P value obtained from x2 test.‡ 27 patients not included due to missing values.§ Only patients without missing variable values included.

Table 3.

Crude Distribution of Periodontal Status by Group

Periodontal

Status

COPD Group Severity

Control Group COPD P* Stage I Stage II Stage III Stage IV P†

Mild 159 (36.3%) 73 (12.6%) 20 (24.7%) 26 (9.8%) 20 (11.6%) 7 (10.9%)

Moderate 86 (19.6%) 265 (45.6%) 43 (53.1%) 159 (60.2%) 57 (33.1%) 6 (9.4%)

Severe 193 (44.1%) 243 (41.8%) 18 (22.2%) 79 (29.9%) 95 (55.2%) 51 (79.7%)

Total 438 (100%) 581 (100%) <0.001 81 (100%) 264 (100%) 172 (100%) 64 (100%) <0.001

* P value obtained from x2 test for comparing the distribution of periodontitis between control and COPD groups.† P value obtained from x2 test for comparing the distribution of periodontitis among different COPD groups.


1291

the relationship between periodontitis andCOPD.

Previous research has suggested anassociation between poor oral hygiene,periodontitis, and chronic respiratory dis-eases such as COPD.12,33 The study ofHayes et al.34 found that the level ofABLassessed by periapical radiographswas independently associated with COPD.Scanapieco and Ho.12 performed a retro-spective study of 13,792 individuals ‡20years old with ‡6 natural teeth; the resultsshowed that individuals with COPD hadhigher AL than those without COPD, andthe individuals with mean AL ‡3.0 mmhad a higher odds of COPD than those withmean AL <3.0 mm (OR = 1.45). In a pro-spective study18 of 1,112 participants fol-lowed up to 30 years, participants in theincreasing quintile of worse periodontalhealth at baseline had greater risk for de-veloping COPD. Some studies reportedthat periodontitis was common in patientswith severe COPD,17,35 and poor peri-odontal health was associated with an in-creased severity of COPD.17 A recentIndian study17 also showed that the highermean AL, the higher likelihood of COPD.Leuckfeld et al.35 reported that mean mar-ginal bone level ‡4 mm was significantlyassociated with severe COPD. We have re-ported previously in a Chinese populationthat patients having less healthy oral be-havior and poor oral health knowledgewere more likely to have COPD.36 Thepresent study expands our previous find-ings to a larger sample and shows consis-tent and strong relationship betweenmultiple periodontal indices and COPDstatus. When compared to participantswith a BODE score of 0, participants withhigher BODE scores had significantlyhigher BI, AL, PI, ABL, and less numberof teeth. Mean of PD, AL, PI, ABL, lessnumber of teeth, percentage of PD ‡4mm and AL ‡3 mm, and percentage ofPD ‡5 mm and AL ‡4 mm significantlyincreased with increasing severity ofCOPD.

In this study, PI was found the mainperiodontal health-related factor forCOPD. Plaque accumulation is clearly anessential initial etiological factor inperiodontitis, although the exact mecha-nism for the relationship between oral hy-giene and COPD remains unclear. SomeT

ab

le4.

Periodonta

lS

tatu

sby

Gro

up

CO

PD

Stag

esBO

DE

Score

s

CO

PD

Gro

ups

BO

DE

>0G

roup

s

Periodont

al

Indic

ator

Cont

rolG

roup

(n=

438)

CO

PD

(n=

581)

P*St

age

I

(n=

81)

Stag

eII

(n=

264)

Stag

eIII

(n=

172)

Stag

eIV

(n=

64)

P†BO

DE

=0

(n=

429)

BOD

E>

0

(n=

590)

P‡BO

DE

=1

or

2(n

=320)

BO

DE

=3

or

4(n

=180)

BO

DE

‡5(n

=90)

P§

PD

(mm

)3.4

5–

0.9

23.5

4–

0.7

10.0

73.4

1–

0.5

53.5

4–

0.6

53.4

2–

0.6

34.0

0–

1.0

5<0

.001

3.5

1–

0.8

73.4

8–

0.7

60.5

83.4

5–

0.7

53.4

6–

0.6

63.6

8–

0.9

40.0

84

BI

2.5

9–

0.6

82.6

0–

0.6

10.8

52.7

2–

0.6

32.5

9–

0.5

62.5

1–

0.6

32.6

9–

0.7

30.0

92.6

6–

0.6

22.5

4–

0.6

50.0

03

2.5

3–

0.6

42.5

5–

0.6

52.5

5–

0.6

90.0

3

PI

2.5

6–

0.4

42.7

9–

0.2

9<0

.001

2.6

6–

0.3

52.7

8–

0.3

02.8

4–

0.2

42.8

8–

0.2

1<0

.001

2.5

8–

0.4

22.7

7–

0.3

2<0

.001

2.7

1–

0.3

62.8

2–

0.2

82.8

8–

0.1

8<0

.001

AL

(mm

)4.8

7–

1.8

35.5

2–

1.5

5<0

.001

5.0

3–

1.2

85.5

4–

1.3

95.5

4–

1.5

25.9

7–

2.2

8<0

.001

4.9

3–

1.7

85.4

6–

1.6

2< 0

.001

5.2

1–

1.5

35.6

6–

1.5

05.9

6–

1.9

7<0

.001

ABL

1.3

0–

0.4

11.4

9–

0.4

5<0

.001

1.3

5–

0.3

11.5

1–

0.4

41.4

7–

0.4

51.7

–0.5

8<0

.001

1.3

2–

0.4

21.4

7–

0.4

5<0

.001

1.4

–0.3

91.5

3–

0.4

71.6

1–

0.5

5<0

.001

Num

ber

of

teet

h24.3

9–

6.0

023.0

2–

6.4

80.0

01

25.2

1–

3.9

323.0

4–

6.2

922.4

1–

6.9

821.7

7–

7.7

7<0

.001

24.5

1–

5.9

422.9

5–

6.4

9<0

.001

24.0

4–

5.5

622.5

2–

6.6

919.9

4–

8.0

1<0

.001

PD

‡5m

man

d

AL

‡4m

m(%

)

18.3

3–

0.2

319.9

0–

0.2

00.2

415.6

1–

0.1

419.4

8–

0.1

716.2

7–

0.1

636.8

4–

0.3

3<0

.001

18.9

1–

0.2

219.4

5–

0.2

00.6

918.3

8–

0.1

918.5

5–

0.1

725.0

9–

0.3

00.0

5

PD

‡4m

man

d

AL

‡3m

m(%

)

36.7

8–

0.3

042.8

1–

0.2

5<0

.001

38.7

5–

0.2

143.9

7–

0.2

338.1

8–

0.2

555.5

8–

0.3

4<0

.001

39.0

2–

0.2

841.0

9–

0.2

70.2

339.7

1–

0.2

640.6

4–

0.2

646.8

9–

0.3

10.1

0

*P

valu

eobta

ined

from

indep

enden

tt

test

for

com

pari

ng

index

esof

per

iodontitis

bet

wee

nco

ntr

oland

CO

PD

gro

up

s.†

Pva

lue

obta

ined

from

one-

way

AN

OV

Afo

rco

mp

ari

ng

index

esof

per

iodontitis

am

ong

diffe

rent

CO

PD

stages

.‡

Pva

lue

obta

ined

from

indep

enden

tt

test

for

com

pari

ng

index

esof

per

iodontitis

bet

wee

nB

OD

E=

0gro

up

and

BO

DE

>0

gro

up

.§

Pva

lue

obta

ined

from

one-

way

AN

OV

Afo

rco

mp

ari

ng

index

esof

per

iodontitis

am

ong

diffe

rent

BO

DE

score

s.


1292

hypotheses suggest that bacteria in oralcavity may be aspirated along with re-spiratory pathogens and affect adhesionof the later organisms to the respiratoryepithelium, which subsequently causelung disease.16 Dental plaque may alsoprovide nutrition to the pathogens in re-spiratory tract, especially in patientswith poor oral hygiene.3 Periodontal dis-ease may alter environmental conditionsto permit mucosal colonization and infec-tion by respiratory pathogens.34,37 Aspi-ration of oral pharyngeal contents, suchas foodparticlesandsaliva rich inbacteria,would link periodontitis and respiratory in-fections.38 Concentration of bacteria in sa-liva is very high, and species of bacteria inoral cavity has been found in the lungs ofpatients with COPD.39 Some trials haveshown that the risk for respiratory diseaseand infection of lower respiratory tractcould be reduced through improving oralhygiene and decreasing the bacteria levelin oral cavity.17,40

It is possible that periodontal condi-tions combine with other factors, suchas smoking, environmental pollutants,allergies, and genetics, to contribute tothe progression of COPD.17 Cigarettesmoking is the most important and awell-accepted risk factor for COPD.12

Smoking is also a major risk factor forperiodontitis. Therefore, smoking couldbe either a confounder or an effectmodifier in the relationship betweenperiodontitis and COPD.41 Katanciket al.42 found that AL was higher inpatients with COPD than in the con-trol group, especially among formersmokers. Hyman et al.41 reported thatcurrent smokers with AL ‡4 mm hada high odds of COPD (OR = 3.71). In thecurrent study, to address the interactionamong cigarette smoking status, peri-odontal disease, and COPD, analysiswas stratified by smoking status. The re-sults showed that periodontal indices (BI,PI, AL, and number of teeth) were signif-icantly associated with BODE scoresamong former smokers, and higher PIwas associated with a high odds of COPDamong current smokers.

The present case-control study is thefirst comprehensive analysis of the asso-ciation between periodontitis and COPDin a Chinese population. The resultsT

ab

le5.

Corr

ela

tions

Betw

een

Periodonta

lS

tatu

sand

Lung

Function

Measure

sA

mong

Patients

With

CO

PD

FEV

1(%

Pre

dic

ted)

FEV

1/F

VC

MR

CSc

ore

6M

WD

BO

DE

Score

Periodont

alSt

atus

Pear

son

Corr

elat

ion

Coef

ficie

ntP

Val

ue*

Pear

son

Corr

elat

ion

Coef

ficie

ntP

Val

ue*

Spea

rman

Corr

elat

ion

Coef

ficie

ntP

Val

ue*

Pear

son

Corr

elat

ion

Coef

ficie

ntP

Val

ue*

Spea

rman

Corr

elat

ion

Coef

ficie

ntP

Val

ue*

PD

(mm

)-0

.081

0.0

1†

-0.0

71

0.0

2†

0.0

46

0.2

80.0

31

0.3

30.0

39

0.2

2

BI

0.0

37

0.2

40.0

21

0.5

1-0

.030

0.3

20.0

20

0.5

2-0

.056

0.0

7

PI

-0.3

45

<0.0

01§

-0.3

54

<0.0

01§

0.2

61

<0.0

01§

0.0

40

0.2

10.2

71

<0.0

01§

AL

(mm

)-0

.182

<0.0

01§

-0.2

07

<0.0

01§

0.1

71

<0.0

01§

-0.0

53

0.0

90.1

91

<0.0

01§

ABL

-0.2

16

<0.0

01§

-0.2

25

<0.0

01§

0.1

61

<0.0

01§

-0.0

88

0.0

06‡

0.2

02

<0.0

01§

Num

ber

of

teet

h0.1

55

<0.0

01§

0.1

29

<0.0

01§

-0.1

36

<0.0

01§

0.1

09

0.0

01‡

-0.1

84

<0.0

01§

PD

‡5m

man

dA

L‡4

mm

(%)

-0.0

90

0.0

04‡

-0.0

68

0.0

3†

0.0

84

0.2

40.0

20

0.5

30.0

58

0.0

6

PD

‡4m

man

dA

L‡3

mm

(%)

-0.1

22

<0.0

01§

-0.1

17

<0.0

01§

0.0

66

0.0

3†

0.0

16

0.6

10.0

75

0.0

2†

*P

valu

eobta

ined

from

biv

ari

ate

Pea

rson

corr

elation

analy

sis.

†£0

.01

;P

<0.0

5.

‡£0

.00

1;

P<0

.01

.§

P<0

.00

1.


1293

indicated the importance of promoting oral hygiene inthe prevention and treatment of both diseases. Suchdata, if confirmed by future prospective studies, wouldprovide valuable information for the planning of oralhealth care for patients with COPD. The authors alsohope that this study will stimulate and support futurefunction studies to offer new insights into the patho-genic mechanisms underlying the relationship of peri-odontal health and COPD.

There are several limitations that deserve to beconsidered. First, Berkson’s bias existed in this study,which is a type of selection bias, and may occur in ret-rospective case-control studies based entirely on hos-pital cases and controls. Second, the causality cannotbe inferred from this study because of its observa-tional and retrospective study design. Third, this studycannot completely exclude the possibility of residualconfounding by other healthy lifestyle variables andcomorbid disease. In addition, measurment errorsof periodontal and pulmonary variables could haveattenuated the results.

CONCLUSIONS

In conclusion, a strong association between peri-odontitis and COPD was found. The results showed

that individuals with higher BODE scores hadsignificantly higher BI, PI, AL, ABL, and less numberof teeth. These findings indicate the potential im-portance of promoting oral hygiene in the preventionand treatment of COPD, although additional studiesare needed to clarify the causal relationship betweenperiodontitis and COPD and explore the biologicmechanisms underlying the observed association.

ACKNOWLEDGMENTS

This study was supported by an International Scienceand Technology Cooperation research grant of Bei-jing Municipal Science and Technology Commission(2006, Beijing, China) and National Natural ScienceFoundation of China Grant 30872878. The fundingagency was Beijing ChaoYang Hospital, affiliated withCapital Medical University, Beijing, China. The au-thors thank all participants in this study. Dr. Yan Siand Hong Fan are co-first authors and contributedequally to this study. The authors report no conflictsof interest related to this study.

REFERENCES1. Calverley PM, Walker P. Chronic obstructive pulmo-

nary disease. Lancet 2003;362:1053-1061.

Table 6.

ORs of COPD in Relation to Periodontal Health in all Participants and Stratified byCigarette Smoking Status

95% CI

Covariates Entered in the Models Adjusted OR Lower Upper P Value*

All participantsSmoking status 0.48 0.27 0.85 0.01†

Total smoking quantity 1.01 1.00 1.01 0.05†

PD 0.19 0.07 0.50 0.001‡

PI 9.01 3.98 20.40 <0.001§

AL 1.41 1.09 1.84 0.01†

Current smokersPI 8.28 2.36 29.00 0.001PD ‡5 mm and AL ‡4 mm (%) 0.16 0.03 0.84 0.03

Former smokersBI 0.23 0.13 0.41 <0.001PI 5.89 2.64 13.12 <0.001§

AL 1.59 1.24 2.02 <0.001§

Number of teeth 1.07 1.02 1.13 0.006‡

Non-smokersPI 2.46 1.47 4.10 0.001‡

ABL 3.89 2.00 7.57 <0.001§

PD ‡5 mm and AL ‡4 mm (%) 0.19 0.06 0.64 0.007‡

BMI 0.83 0.78 0.88 <0.001§

* P value obtained from logistic regression analysis (with age, sex, occupation, educational level, and smoking status adjusted).† P <0.05.‡ 0.001£ P <0.01.§ P <0.001.


1294

2. Imfeld S, Bloch KE, Weder W, Russi EW. The BODEindex after lung volume reduction surgery correlateswith survival. Chest 2006;129:873-878.

3. Gunen H, Hacievliyagil SS, Yetkin O, Gulbas G, MutluLC, Pehlivan E. Prevalence of COPD: First epidemio-logical study of a large region in Turkey. Eur J InternMed 2008;19:499-504.

4. Lopez AD, Shibuya K, Rao C, et al. Chronic obstruc-tive pulmonary disease: Current burden and futureprojections. Eur Respir J 2006;27:397-412.

5. Taylor JD. COPD and the response of the lung totobacco smoke exposure. Pulm Pharmacol Ther 2010;23:376-383.

6. Rabe KF, Hurd S, Anzueto A, et al. Global strategyfor the diagnosis, management, and prevention ofchronic obstructive pulmonary disease: GOLD execu-tive summary. Am J Respir Crit Care Med 2007;176:532-555.

7. Brown LJ, Loe H. Prevalence, extent, severity andprogression of periodontal disease. Periodontol 20001993;2:57-71.

8. Qi XQ. Investigation Report of the Third National OralHealth Survey in China [in Chinese], Beijing: People’sMedical Publishing House;2008:69-111.

9. Slade GD, Ghezzi EM, Heiss G, Beck JD, Riche E,Offenbacher S. Relationship between periodontal dis-ease and C-reactive protein among adults in theAtherosclerosis Risk in Communities Study. ArchIntern Med 2003;163:1172-1179.

10. Noack B, Genco RJ, Trevisan M, Grossi S, Zambon JJ,De Nardin E. Periodontal infections contribute toelevated systemic C-reactive protein level. J Peri-odontol 2001;72:1221-1227.

11. Kuo LC, Polson AM, Kang T. Associations betweenperiodontal diseases and systemic diseases: a reviewof the inter-relationships and interactions with di-abetes, respiratory diseases, cardiovascular dis-eases and osteoporosis. Public Health 2008;122:417-433.

12. Scannapieco FA, Ho AW. Potential associations be-tween chronic respiratory disease and periodontaldisease: Analysis of National Health and NutritionExamination Survey III. J Periodontol 2001;72:50-56.

13. Genco RJ, Grossi SG, Ho A, Nishimura F, MurayamaY. A proposed model linking inflammation to obesity,diabetes, and periodontal infections. J Periodontol2005;76(Suppl. 11):2075-2084.

14. Scannapieco FA, Genco RJ. Association of periodon-tal infections with atherosclerotic and pulmonarydiseases. J Periodontal Res 1999;34:340-345.

15. Scannapieco FA, Papandonatos GD, Dunford RG.Associations between oral conditions and respiratorydisease in a national sample survey population. AnnPeriodontol 1998;3:251-256.

16. Scannapieco FA, Wang BY, Shiau HJ. Oral bacteriaand respiratory infection: Effects on respiratorypathogen adhesion and epithelial cell proinflamma-tory cytokine production. Ann Periodontol 2001;6:78-86.

17. Deo V, Bhongade ML, Ansari S, Chavan RS. Peri-odontitis as a potential risk factor for chronic obstruc-tive pulmonary disease: A retrospective study. IndianJ Dent Res 2009;20:466-470.

18. Garcia RI, Nunn ME, Vokonas PS. Epidemiologicassociations between periodontal disease and chronic

obstructive pulmonary disease. Ann Periodontol 2001;6:71-77.

19. Fletcher CM. Standardized questionnaire on respira-tory symptoms. BMJ 1960;2:1665.

20. Muhlemann HR, Son S. Gingival sulcus bleeding. Aleading symptom in initial gingivitis. Helv OdontolActa 1971;15:107-113.

21. Silness J, Loe H. Periodontal disease in pregnancy.II. Correlation between oral hygiene and periodontalcondition. Acta Odontol Scand 1964;22:121-135.

22. Meng HX. Periodontology [in Chinese], 3th ed. Beijing:People’s Medical Publishing House;2008:117-136.

23. Albandar JM, Brunelle JA, Kingman A. Destructiveperiodontal disease in adults 30 years of age and olderin the United States, 1988-1994. J Periodontol 1999;70:13-29.

24. Stavem K, Sandvik L, Erikssen J. Can global initiativefor Chronic Obstructive Lung Disease stage 0 provideprognostic information on long-term mortality in men?Chest 2006;130:318-325.

25. Huijsmans RJ, de Haan A, ten Hacken NN,Straver RV, van’t Hul AJ. The clinical utility of theGOLD classification of COPD disease severity inpulmonary rehabilitation. Respir Med 2008;102:162-171.

26. Rasekaba T, Lee AL, Naughton MT, Williams TJ,Holland AE. The six-minute walk test: A useful metricfor the cardiopulmonary patient. Intern Med J 2009;39:495-501.

27. Bestall JC, Paul EA, Garrod R, Garnham R, Jones PW,Wedzicha JA. Usefulness of the Medical ResearchCouncil (MRC) dyspnoea scale as a measure ofdisability in patients with chronic obstructive pulmo-nary disease. Thorax 1999;54:581-586.

28. Nishimura K, Izumi T, Tsukino M, Oga T. Dyspnea is abetter predictor of 5-year survival than airway ob-struction in patients with COPD. Chest 2002;121:1434-1440.

29. Celli BR, Cote CG, Marin JM, et al. The body-massindex, airflow obstruction, dyspnea, and exercisecapacity index in chronic obstructive pulmonary dis-ease. N Engl J Med 2004;350:1005-1012.

30. Faganello MM, Tanni SE, Sanchez FF, Pelegrino NR,Lucheta PA, Godoy I. BODE index and GOLD stagingas predictors of 1-year exacerbation risk in chronicobstructive pulmonary disease. Am J Med Sci 2010;339:10-14.

31. Celli BR. Predictors of mortality in COPD. Respir Med2010;104:773-779.

32. Liu SF, Chin CH, Wang CC, Lin MC. Correlationbetween serum biomarkers and BODE index in pa-tients with stable COPD. Respirology 2009;14:999-1004.

33. Scannapieco FA, Bush RB, Paju S. Associationsbetween periodontal disease and risk for nosocomialbacterial pneumonia and chronic obstructive pulmo-nary disease. A systematic review. Ann Periodontol2003;8:54-69.

34. Hayes C, Sparrow D, Cohen M, Vokonas P, Garcia RI.The association between alveolar bone loss andpulmonary function: The VA Longitudinal Study. AnnPeriodontol 1998;3:257-261.

35. Leuckfeld I, Obregon-Whittle MV, Lund MB, GeiranO, Bjørtuft O, Olsen I. Severe chronic obstructivepulmonary disease: Association with marginal boneloss in periodontitis. Respir Med 2008;102:488-494.


1295

36. Wang Z, Zhou X, Zhang J, et al. Periodontal health,oral health behaviours, and chronic obstructive pul-monary disease. J Clin Periodontol 2009;36:750-755.

37. Scannapieco FA, Mylotte JM. Relationships betweenperiodontal disease and bacterial pneumonia. J Peri-odontol 1996;67(Suppl. 10):1114-1122.

38. Page RC. Periodontitis and respiratory diseases: Dis-cussion, conclusions, and recommendations. AnnPeriodontol 2001;6:87-90.

39. Terpenning MS. The relationship between infectionsand chronic respiratory diseases: An overview. AnnPeriodontol 2001;6:66-70.

40. Fourrier F, Cau-Pottier E, Boutigny H, Roussel-Delvallez M, Jourdain M, Chopin C. Effects of dentalplaque antiseptic decontamination on bacterial colo-

nization and nosocomial infections in critically illpatients. Intensive Care Med 2000;26:1239-1247.

41. Hyman JJ, Reid BC. Cigarette smoking, periodontaldisease, and chronic obstructive pulmonary disease.J Periodontol 2004;75:9-15.

42. Katancik JA, Kritchevsky S, Weyant RJ, et al. Peri-odontitis and airway obstruction. J Periodontol 2005;76(Suppl. 11):2161-2167.

Correspondence: Zuomin Wang, Beijing ChaoYang Hospi-tal, Capital Medical University, Chao Yang District, Beijing,China, 100020.

Submitted August 11, 2011; accepted for publicationDecember 15, 2011.


1296

Documents

Association Between Periodontitis and Chronic Obstructive Pulmonary Disease in a Chinese Population