UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Prevalence and progression of untreated periodontal disease in a youngIndonesian population

Timmerman, M.F.

Publication date2001

Link to publication

Citation for published version (APA):Timmerman, M. F. (2001). Prevalence and progression of untreated periodontal disease in ayoung Indonesian population.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an opencontent license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, pleaselet the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the materialinaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letterto: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. Youwill be contacted as soon as possible.

Download date:06 Sep 2021

UNTREATED PERIODONTAL DISEASE IN INDONESIAN ADOLESCENTS Longitudinal clinical data & prospective clinical and microbiological risk assessment

CHAPTER 4

Timmerman, M.F.1

Van der Weijden G.A.1

Abbas, F.1

Arief, E.M.2

Armand, S.2

Winkel E.G.1

Van Winkelhoff, A.J . 3

Van der Velden, U.1

Department of Periodontology, Academic Centre for Dentistry Amsterdam, The Netherlands Department of Periodontology, Padjadjaran State University, Bandung, Indonesia Department of Oral Microbiology, Academic Centre for Dentistry Amsterdam, The Netherlands

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

Epidemiologic surveys conducted throughout the world point to the almost

universal distribution destructive periodontal diseases (Polsky 1996), although

the geographic distribution varies widely. For instance, some populations on the

Indian subcontinent as well as in other parts of Asia seem to be more

susceptible to destructive periodontal disease than North Americans or

Europeans. Studies of natives in the Pacific area (Davies 1956), Africa, (Olsson

1978), and South America (Russell & Ayers 1960) indicate that the prevalence

in some populations is conspicuously high and that the severity may vary

considerably within subpopulation (Löe & Morrison 1990).

In the cross-sectional studies of the late fifties and early sixties, plaque and

calculus were shown to play a major rôle in the etiology of periodontal disease

(Mobley & Smith 1963, Ladavalya & Harris 1959, Schei et al. 1959, Lövdal et

al. 1958). Since then, the knowledge has markedly increased regarding the

etiology of periodontal disease. Intense study of specific risk factors for

periodontal diseases has pointed to a few members of the periodontal

microbiota as candidate pathogens for initiation and progression of periodontal

disease. Grossi et al. (1994, 1995), in a cross-sectional study, tested a panel of

micro-organisms including Actinobacillus actinomycetemcomitans, Bacteroides

forsythus, Campylobacter rectus, Capnocytophaga species, Eubacterium

sabureum, Fusobacterium nucleatum, Porphyromonas gingivalis, and Prevotella

intermedia. Of these organisms P. gingivalis and B. forsythus were associated

with increased risk for attachment loss as a measure of periodontal disease after

adjustment for age, plaque, smoking and diabetes.

Although cross-sectional studies of clinical and microbiological factors can be

meaningful, longitudinal studies are needed to study the natural history of

disease and its determinants. Knowledge of the natural history allows analysis

of potential factors and conditions that may have an impact on the disease

process, possibly resulting in the development of effective control measures

(Löe & Morrison 1990). Such knowledge will facilitate (i) the identification of

those individuals for whom periodontal disease will become a serious oral health

problem, prior to the development of irreversible periodontal damage, and (ii) the

application of effective preventive measures (Papapanou 1993).

7 5

CHAPTER 4

A longitudinal study was initiated in 1987 in a young population, which had not

received regular dental care, in order to establish possible clinical and

microbiological risk markers for periodontitis (Timmerman et al. 1998). The

material presented in this paper describes the clinical periodontal condition at

baseline (1987) and after a period of 7 years (1994). Furthermore, the

relationship was assessed between progression of the disease and baseline

clinical and microbiological data.

MATERIALS AND METHODS

Study population

For this longitudinal, prospective study a village with approximately 2.000

inhabitants at the Malabar/Poerbasari tea estate on Western Java, Indonesia

was selected. At the baseline evaluation in 1987 (Timmerman et al. 1998) all

subjects (n = 255) in the age range 1 5 - 2 5 years participated in this

investigation. This specific population was chosen because it had not received

regular dental care and had not been exposed to preventive dental care

programs. Emergency dental treatment, consisting of extraction of teeth was

provided by a general physician. Therefore, this population was suitable to

study the natural development and progression of periodontitis.

The population consisted mostly of tea labourers with a low educational level

employed by a government owned tea estate, PTP XIII. Most individuals had

received a limited or complete primary education (N = 204) whereas 51

individuals completed further education. The subjects received basic medical

care and there was no indication of obvious malnutrition.

Examination procedures

At the baseline examination participants were asked about their educational

level, general health status and recent use of antibiotics. In addition, family

relationships were recorded (Van der Velden et al. 1993) prior to taking the

clinical measurements. Subsequently, all samples for bacteriological examination

were obtained except those from the pockets. The clinical assessments were

carried out by 3 periodontists (FA, SA, EGW) and dictated to a chairside

assistant. Every 10th subject was used for double scores of the clinical

76

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

parameters located at the Ramfjord teeth, which were carried out at the end of

each morning and afternoon session. The clinical examination was performed in

an office facility of the tea factory and portable dental chairs were used. After

clinical measurements were completed, samples from the selected pockets were

taken for bacteriological examination.

At the follow up assessment in 1994 the clinical examination was performed in

the same way and family relationships were verified. In a randomly chosen

sample of 18 subjects all clinical measurements were repeated by each of the

three examiners (FA, EGW, GAW) in each patient to establish the inter-examiner

variance.

Clinical Parameters

The following indices were recorded:

Plaque (Silness & Lóe 1964)

Calculus (Björby & Löe 1967)

Pocket Depth (PD) using a force controlled probe (Brodontic® Ash/

Dentsply, 240 N/cm2) with a Williams calibration.

Bleeding on Probing (BOP; Van der Velden 1979) using the force

controlled probe using a three point scale: 0) non-bleeding sites;

1 ) 'pin prick' bleeding;

2) 'excess' bleeding.

Bleeding was scored within 30 seconds after probing.

Attachment loss (AL) assessed by subtracting the distance between the

gingival margin and the cemento-enamel junction (GM-CEJ) from the

recorded probing depth or, in case of gingival recession adding the GM-

CEJ value to the probing depth measurement. The GM-CEJ distance was

evaluated by means of a Hu-Friedy® probe (Williams calibration).

The sequence of scoring was always the same. Clinical parameters were scored

on the approximal surfaces of the vestibular aspect of all teeth except third

molars. Calculus was scored on the approximal surfaces of the Ramfjord teeth

(16, 2 1 , 24, 36, 41 and 44) (Ramfjord 1959). Measurements were rounded off

to the nearest millimetre.

77

CHAPTER 4

Microbiological examination

At the baseline evaluation, samples for bacteriological examination were taken

from various parts of the oral cavity and were collected in the following order:

the dorsum of the tongue, from the vallate papillae to the tip of the

tongue

the buccal gingiva in the upper jaw, from the left to the right first molar

the saliva

the deepest bleeding pocket without attachment loss (Timmerman et al.

1998).

Samples from the tongue and the gingiva were obtained by sweeping a sterile

swab under continuous pressure over the total surface. In case of the gingiva,

care was taken not to disturb the supragingival plaque. The sample from the

tongue was suspended in 1.8 ml reduced transport fluid supplemented with

10% Fildes extract (RTFF, Petit et al. 1991). The gingival sample was

suspended in 0.9 ml RTFF. Saliva was sampled by adding approximately 1 ml of

unstimulated saliva to 0.9 ml RTFF. After all clinical measurements were

completed the deepest bleeding pocket without clinical loss of attachment was

selected and sampled (the sampled pocket, SP, réf. Timmerman et al. 1998).

The microbiological data on a pocket showing all characteristics of periodontitis

except probing attachment loss may serve as a risk marker for onset and

progression of periodontal disease. Furthermore, the choice of a pocket without

attachment loss enabled an evaluation of the subgingival flora of a comparable

type of pocket in all subjects at the time of the baseline measurements.

After careful removal of the supragingival plaque with a curette a subgingival

sample was taken, using a nerve broach (Maillefer®) wound with cotton and

heat sterilized. The part of the nerve broach that had been inserted was cut off

and suspended in 0.9 ml RTFF (Van Winkelhoff et al. 1988). Specimens were

vortexed for 60 sec. at the maximum setting. Samples were processed for

phase contrast microscopy to establish the prevalence of spirochetes (Spir) and

motile micro-organisms (Mot MO). Furthermore, for indirect immunofluorescence

examination to establish presence of A. actinomycetemcomitans {Aa), P.

gingivalis (Pg), P. intermedia (Pi), the samples were fixed with formaldehyde

(0.2%, v/v). Ten /J\ aliquots of the sample were transferred to multi-well slides,

air-dried and gently heat-fixed. Slides were stored at room temperature until

transportation to The Netherlands. Upon arrival in Amsterdam, slides were

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

stored at -20°C until further processing for indirect immunofluorescence assay

could take place (Timmerman et al. 1998).

Data analysis

For both the baseline and follow-up evaluations the clinical parameters assessed

at all approximal surfaces were calculated as mean scores per patient. The

number of sites showing attachment loss > 2 mm and the number of sites

showing probing depth of > 5 mm were enumerated. Calculus scores were

analyzed as the number of approximal surfaces of the Ramfjord teeth showing

subgingival calculus.

For each patient, data concerning probing depth and attachment loss at baseline

and follow-up were pooled at a site level. Differences between baseline and

follow-up were calculated at site level. Using a difference of > 2 mm, a site was

accepted as showing change in pocket depth or attachment level, as was

suggested by Haffajee et al. (1983) and Lindhe et al. (1993). Subjects were

defined as Progressive Disease Subjects (PDS), if they showed > 1 site that lost

attachment > 2 mm. The other subjects were defined as Non-Progressive

Disease Subjects (NPDS) (Baelum et al. 1997).

Microbiological data at baseline were analyzed for the prevalence of the micro­

organisms included in the study by establishing the proportion of subjects

positive for the various micro-organisms. Comparisons between the different

locations of sampling were made using a Friedman test.

Inter-examiner error was estimated by calculation of the inter-examiner variance

using a repeated measures model of the analysis of variance both at site level

(at baseline; Plaque: 0.25, Calculus: 0.32, BOP: 0.30, PD: 0.23, AL: 0.37; at

follow-up; Plaque: 0.32, Calculus: 0.44, BOP: 0.38, PD: 0.35, AL: 0.77) and at

patient mean level (at baseline; Plaque: 0.07, Calculus: 0.05 BOP: 0.09, PD:

0.06, AL: 0.07; at follow-up; Plaque: 0.07, Calculus: 0.05 BOP: 0.09, PD:

0.06, AL: 0.07). These data fall well within the boundaries of reproducibility as

described by Haffajee and Socransky (1986).

Comparisons of clinical parameters between baseline and follow-up were made

using a Wilcoxon signed rank test. Comparisons of clinical parameters between

PDS and NPDS were made using a Mann-Whitney test. Odds ratios were

calculated for progressive disease (NPDS versus PDS) with prevalence of the

various micro-organisms at the different sampling sites.

79

CHAPTER 4

The calculation of a predictive value of the various background, clinical and

microbiological parameters at baseline, was obtained using a stepwise logistic

regression model analyzing the association of progressive disease (NPDS versus

PDS) with age, gender, mean pocket depth, mean pre-existent attachment loss,

bleeding on probing, plaque index, the number approximal sites on Ramfjord

teeth showing subgingival calculus (score 2 and 3), number of sites with PD > 5

mm, number of sites w i th AL > 2 mm, prevalence of A.

actinomycetemcomitans, P. gingivalis, P. intermedia, spirochetes and motile

micro-organisms in the sampled pocket, on the gingiva, on the tongue and in

the saliva.

In order to study local factors to explain local disease activity each subject was

characterized in a constrained design by the 'Sampled Pocket' (SP) only. Data

on the SP were analyzed as a single response site per patient.

If the SP showed attachment loss > 2 mm, subjects were defined as

Progressive Disease Subjects in the Sampled Pocket (PDS-SP). The other

subjects were defined as Non-Progressive Disease Subjects in the Sampled

Pocket (NPDS-SP). Differences between clinical parameters for PDS-SP and

NPDS-SP were analyzed using a Mann-Whitney test. Comparisons between

baseline and follow-up were made using a Wilcoxon signed rank test. Odds

ratios were calculated for progressive disease (NPDS-SP versus PDS-SP) with

prevalence of the various micro-organisms.

In order to clarify the pattern of association of disease progression with the

various background, clinical and microbiological parameters of the SP at

baseline, associations were analyzed using a path analysis, establishing relations

between all entered parameters in three levels, by using stepwise logistic

regression and stepwise multiple linear regression models.

At the first level a stepwise logistic regression model was used to analyze the

association of progressive disease (NPDS-SP versus PDS-SP) with age, gender,

baseline probing pocket depth, bleeding on probing, plaque index, and

prevalence of A. actinomycetemcomitans, P. gingivalis, P. intermedia,

spirochetes and motile micro-organisms in the SP. Parameters that showed

significant association (main factors), were submitted to the next 2 levels of

analysis, exploring the relationship of subsidiary factors with the main factors,

using both logistic and multiple linear regression, where appropriate.

Values of p < 0.05 were accepted as statistically significant.

80

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

RESULTS

Clinical results

At baseline, the study population consisted of 130 males and 125 females with

a mean age of 20.0 yrs (SD 3.2). At the follow-up assessment, 167 of these

subjects (75 male, 92 female) of the original group of 255 were available for

evaluation. The other 88 persons could not be retrieved, because they had

moved away from the village. Of these 167 subjects, data of 160 were

analyzed in the present study. The other 7 subjects were excluded from the

analysis because all approximal sites showed loss of attachment at the baseline

examination in 1987, which made it impossible to study the composition of the

subgingival microbiota of a site without attachment loss.

In Table 1 the clinical data are presented as mean values for the entire

population as well as for the subgroups Progressive Disease Subjects (PDS) and

the Non-Progressive Disease Subjects (NPDS). The mean number of teeth

present per subject was 27.4 at baseline and 26.9 at follow-up. The mean loss

of attachment at baseline was 0.33 mm, which increased with a statistically

significant increment to 0.73 mm at follow-up (p < 0.00005). At baseline the

mean pocket depth was 3.26 mm. At follow-up no significant change could be

found for the probing depth, which was 3.32 mm at that time.

At baseline no difference was observed between males and females for any of

the clinical parameters. At follow-up only the plaque score for men was lower

than the score for women (1.07 and 1.22 respectively p = 0.021).

The group of NPDS consisted of 30 subjects, whereas the remaining 130

subjects were PDS. The mean loss of attachment for the NPDS and PDS was

0.24 mm and 0.35 mm at baseline and 0.33 mm and 0.82 mm at follow-up

respectively. Analysis showed a statistically significant difference between

NPDS and PDS both at baseline and follow-up (p = 0.012 and p < 0.00005,

respectively). In addition, the increment of mean loss of attachment during this

period showed a statistically significant difference between NPDS and PDS (p <

0.00005).

The mean pocket depth was 3.21 mm and 3.27 mm at baseline and 3.13 mm

and 3.37 mm at follow-up for the NPDS and PDS respectively. The pocket

depth at follow-up showed a statistically significant difference between NPDS

and PDS (p = 0.0190). In the PDS the mean number of sites losing attachment

81

co a 0-

Q. o Q.

CT) CT)

co

CT)

CO Û Q .

S^ 5-S ^ o o

co ^ r -J- N- C\J o

o o - j -o o - j ->o fO o < - < - o o o

,— co S-

CO o ro r~-ro r\i CC IS.

o o o o OJ C\J

ro ro • - ro

O LTï O O r o -o

r o o ^ o - i w

O IA -J- L/1

O co r o »o ~ CD

O O o o ro f -

r\i r o r o r -

o i n 00 o

r o o * - T— m L

PJ o , - s ^

i n r o r o -J- CO <-

o o o o r\j r--

Is- u-t r\j -vj-r\i r o co o r o « -

vT <t

^o r*-co R

fD CU E — CJ ro

ro o CD *— o in

o o o CD ro ro

i o o o - f m

«- O O O CD f\l O

D CJ CO t/i O tf o O CA

CD IA a . CO

ro -C tu u

c CA CU

c <u

c tu

X tu

~C0 CJ -C tu tu

2 c

> ro C a o CU ai

O) £_ JÛ J D CA c D

Pd CL CA CA o> Ü 0) 3

-Û c D c c CA i n Ol CU a CO c c C u

x : Al o CU tu CD C O o C o

i Q. ro O X] T) TJ r o

CA o (A IA I J CA l_ fU CU tl) a . ro TD O D_ o c C C

_c X o_ > co CO co C <L> CA « CA O

D_ O ~ö "O t \ j CU c C ro

"O tu en o o O TJ tu E c o c c

-C Ü) [_ ro CD Ol o <U T : 3 tu 0) O •Q CD <ll a~-Q -2 _C C t / ï [/) C/Ï u tu a E E E o ro o 3 D 3 •K 0. < ca o . ^ ̂ ̂ ^ C/1 * * O

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

> 2 mm between 1987 and 1994, was 6.2. Fig. 1 shows the frequency

distribution of individuals according to the number of sites per person with > 2

mm attachment loss between baseline and follow-up. In the group of subjects

that showed > 1 site with attachment loss > 2 mm during the evaluation period,

the majority had > 4 sites with this amount of attachment loss.

Fig. 2 shows for each tooth type the number of subjects with > 2 mm

attachment loss between baseline and follow-up at each particular tooth. This

bar chart illustrates that of all teeth, the incisors in the lower jaw were most

frequently affected, followed by the first molars and the incisors in the upper

jaw.

At baseline, 8 subjects showed visual gingival recession at approximal sites;

mean root exposure 1.19 mm, range 1 - 2 mm. In these 8 subjects the mean

number of teeth showing recession was 1.3. At follow up the number of

subjects with visual gingival recession had increased to 33; mean root exposure

1.41 mm, range 1 - 5 mm. The mean number of teeth showing recession was

1.9. Only one of these subjects belonged to the NPDS-group. There were 2 (out

of the 8) subjects that showed no increase of recession between baseline and

o <D_

CD

E

5 10 15 20 25 30 35 40

number of sites

Fig. 1. Frequency distribution of individuals according to the number of sites per person with

>2 mm attachment loss between baseline and follow-up.

8 3

CHAPTER 4

follow up. The clinical characteristics of the 'Sampled Pocket' (SP) are shown in

Table 2. In 5 of the 160 subjects it was not possible to assess the attachment

level at the follow-up evaluation, resulting in a total of 155 subjects with an

évaluable site. Of these 155 sites, 39 showed attachment loss over the

evaluation period. The mean pocket depth at baseline was 3.68 mm and 3.65

mm at follow-up. No difference in pocket depth between PDS-SP and NPDS-SP

was found at baseline. There was a statistically significant difference at follow-

up (PD 4.79 mm for the PDS-SP and 3.27 mm for the NPDS-SP respectively, p

< 0.00005). At baseline, attachment loss was 0.0 mm and remained 0.0 in the

NPDS-SP (all by definition). Attachment loss increased to 2.38 mm in the PDS-

SP at follow-up. Bleeding scores did not differ at baseline and were 1.16 and

tooth number 17 16 15 14 13 12 11 21 22 23 24 25 26 27

47 46 45 44 43 42 41 31 32 33 34 35 36 37 tooth number

Fig. 2. Number of subjects w i th > 2 mm attachment loss between baseline and fo l low-up at

each particular too th .

8 4

CL CO

<h a CL

CD CD

O 15.

•O t

CO 0) 1 — n

~ C/) <D LO

f 1 0) CL

CL

E

co co -ï r̂ n o o o o

</i CO O 00 O LTi r>

CL ro M O M - - ' -

CQ

o CO CO

^3-

o «-o o o o o X 5 ON

L/t

w o r g ^ ^

<— «- o o o

rn o r\i »- .—

4 J r (1) OJ • — — o (1 o O C ( XJ C c c F m i Dl co

> (1) co •— (1) n - ] "D oo co i : V (1) n

r <li (D 03 * o CL <£. CJ co CL O -fc -X *

._ (IJ CD <n

u o> (fl c Ci­

t o CT) en c 3 U-i

o

-* Q . Q_

t > ^ OJ O

XI c (0

XJ ca E ro Q. XJ

CO ft) N +•» co t o (D o Q

CL D U) U 0J c C C/5 ÛJ OJ OJ o

x : 3 3 IA

C QJ <u co OJ _Q J3 3

CO (/) Q . QJ OJ C

O u OJ --N C L. E V QJ OJ -C

C QJ o (D

CHAPTER 4

1.72 for NPDS-SP and PDS-SP at follow-up respectively. Plaque scores were

different between NPDS-SP and PDS-SP at both assessments. Within these

groups no changes were found over time (1.08 to 1.15 for the NPDS-SP and

1.51 to 1.59 for PDS-SP, p = 0.0026, p = 0.0156, at baseline and follow-up

respectively).

Microbiological results

Table 3 shows the prevalence of A. actinomycetemcomitans, P. gingiva/is, P.

intermedia, spirochetes and motile rods on the tongue, gingiva, in saliva, and in

the sampled pocket. It can be seen that except for A. actinomycetemcomitans

the prevalence of these bacteria was over 88%. Spirochetes and P. intermedia

were the most frequently detected micro-organisms (100%). The highest

prevalence within the oral cavity for A. actinomycetemcomitans was 34% in

the pocket, for P. gingiva/is 66% and 63% respectively, in the pocket and on

the tongue, for P. intermedia 100% on the tongue, for spirochetes 62% in the

pocket and for motile micro-organisms 9 1 % and 89% respectively, on the

tongue and in the saliva.

Table 3. Prevalence of micro-organisms (indirect immunofluorescence and phase contrast) in

sampled sites at baseline (1987)

n = 160 Pocket G i ng i va Tongue Saliva All sites

A. actinomycetemcomitans 34% 167. 217. 107. 537. P. gingivalis 667. 357. 637. 497. 887. P. intermedia B17. 637. 1007. 837. 1007. Spirochetes 627. 137. 467. 537. 897. Motile micro-organisms 597. 197. 917. 897. 1007.

In Table 4 the odds ratios are presented for progressive disease (NPDS versus

PDS) associated with the presence of A. actinomycetemcomitans, P. gingivalis,

P. intermedia, spirochetes and motile micro-organisms on various sampling sites.

For the mucosal sites (tongue, gingiva) and the saliva, no significant odds ratios

were found between the presence of any of the studied micro-organisms and

loss of attachment over the study period.

For the microbiological data of the sampled pocket, an odds ratio for disease

progression at the full mouth level of 4.2 was found for the presence of A.

actinomycetemcomitans and 2.3 for P. gingivalis.

8 6

Q.

•D

en

£ 'E

o (ƒ)

£ 9

cc m u n

w en o

•a Q

o o ~ 1 ^ co CT)

CO (D CO n CD

Q_

LL

cfl

^ u CD

« J J 1R - 1

h- (/)

ro nj o *- L 1 O -J-

r\i co -d- r\J *— co ^J- f\J i— CO

O «— ->J N- r\J O LJ-l - j - * - . r\j f \ i o CO r-

O i - CO CO r\J •— CO S i - \ Û C O co r- r - co

- >d- h - F- . Ï 5 CD

•O -O O « - • co r- r - co «-

1^- K l O O O O o o o *— •o r-*i r o co K I CM o co

3-5 N- ^ CO <=> O O PO

C\J r - « - •ö *— u-1 CO O o o o o

5^ i-n ^ ! LA o ^ J - ^ o S - • r\j co - J co

r o N- f \ i O (M CO N r o «O r- -J" N- r- co i - -O S-

S^ od in >c [> in • O ( M S N - J

—' Q 00 Q —« o . a CL O ; <£ ^ a . ü-S o

IA 00 N O •

—• a co o — ' Q- O CL OL < Z CL S ^ O

— ' Q t / 1 O — ' Q . Q a . ac < z ü . c > f O

> TI (1)

LL

Tl (f)

(D a.

QJ CJ

a i a i

</) co

=> O .

5 <? o CO

I s

1 o CD dî U)

CD -Q 1Q

oT 3

C/5 CD

c 00 t» "aï *̂ ro

Q. E

t =>

CL CO

Q

CL CL E

0 "O 0 E Q. E O

ce 03 O CO '"5

O -Q

c ra

Q l/l

l/l

> i_ — O) w o l/l L_ cu o. O) C o <u L. Cl)

CL. 3 t_ Cl) O . O

C O o

ra ra — a vi Q —• a. a a. ce. < Z CX ï ^ O

G)

Ol O)

C/l C/1

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

The odds ratios for the presence of A. actinomycetemcomitans, P. gingivalis, P.

intermedia, spirochetes and motile micro-organisms with respect to disease

progression in the sampled pocket of NPDS-SP and PDS-SP are presented in

Table 5. For the presence of P. gingivalis the odds ratio was 3.0 and for motile

micro-organisms 3.4. For the other micro-organisms no significant odds ratios

were found.

Logistic regression and path analysis

Table 6 shows the results of the stepwise logistic regression analysis for

progressive disease at the full mouth level with the background parameters, the

clinical parameters and the presence of the various micro-organisms on the

various sampling sites. The factors associated with the progression of disease

were age (p = 0.0417, OR = 1.15), presence of A. actinomycetemcomitans {p

= 0.0087, OR = 4.61) and the number of sites on Ramfjord teeth with

subgingival calculus (p = 0.0214, OR = 1.20). The odds for progression of

disease significantly increased with age and with the amount of subgingival

calculus as well as with the presence of A. actinomycetemcomitans.

Fig. 3 shows the results of the stepwise logistic regression and the subsequent

path analysis with respect to the NPDS-SP and PDS-SP. The primary factors

associated with the progression of disease at the sampled pocket were the

presence of motile micro-organisms (p = 0.0037, OR = 2.78) and the plaque

index at the sampled site at baseline (p = 0.0066, OR = 1.96). At the second

level, the presence of motile micro-organisms was significantly associated with

the presence of P. gingivalis and P. intermedia (p < 0 .001, OR = 10.14, p =

0.002, OR = 4.78, respectively), increasing the odds of motile micro-organisms

being present. The plaque index was significantly related to P. intermedia, i.e.

the plaque index was higher in presence of P. intermedia (p = 0.006, RC =

0.439). At the 3rd level P. gingivalis was associated with the presence of A.

actinomycetemcomitans, where in presence of A. actinomycetemcomitans the

odds of finding P. gingivalis were higher (p < 0.001, OR = 4.80). At this level,

the odds for the presence of P. intermedia were higher in presence of A.

actinomycetemcomitans as well as in presence of P. gingivalis (p < 0.04, OR =

3.42, p = 0.04, OR = 2.43, respectively).

89

CHAPTER 4

Table 6. Results of logistic regression, showing p-values and odds ratios for Progressive Disease

at Full Mouth Level between baseline (1987) and fo l low-up (1994) , Non-Progressive Disease

Subjects (NPDS) versus Progressive Disease Subjects (PDS), associated wi th all risk markers

Risk Markers p-value

Background Markers Age 1.15* Gender

Clinical Markers Pocket depth Attachment loss Bleeding on probing Plaque index Number of sites with subgingival calculus o 1.20* Number of sites with PD^5mm Number of sites AL^2mm

0.04* 0.94

0.47 0.49 0.33 0.81 0.02* 0.97 0.97

Microbiological Markers In the sampled pocket

A. actinomycetemcomitans P. gingival is P. intermedia Spirochetes Motile micro-organisms

On the gingiva A. actinomycetemcomitans P. gingivalis P. intermedia Spi rochetes Motile micro-organisms

On the tongue A. actinomycetemcomitans P. gingivalis P. intermedia Spi rochetes Motile micro-organisms

In the saliva A. actinomycetemcomitans P. gingivalis P. intermedia Spi rochetes Motile micro-organisms

0.01* 0.50 0.93 0.89 0.44

0.22 0.28 0.37 0.95 0.96

0.81 0.85

0.35 0.14

0.86 0.95 0.75 0.62 0.82

* Significant association. ** Per year. *** Per Site. o Score 2 and 3 of Calculus Index as part of Retention Index as scored on approximal surfaces of

Ramfjord teeth (Björby & Löe 1967).

90

« S! 1 O . Q .

Q - T- CD

OJ O > ,13

^ c o ^ i— L. O 'r- •— o QJ T3 OJ "3 > ' i - [/) C3 Û) > £-o j Ta W L. U OJ X J

(U _ -i - vs ro t/1

"O -Q m - a > -a -Q c s 0 TJ • t_ s c\J LO a : O Q. m to

-C\J o

<u a . > U T J -i O QJ : o. —- . i a i ~a E . QJ CO t

— t / î

r; -a Q. +-1 cl] 1=

a i E M - a j na

"S a . S - ë !

c ai ^)f~~ Xs w

£<£ 03 OJ -M O CD

Q —' O —• "3 Öl'r— Q. O û. < O û . ' C Û . Q . w E

O CL

CL

(/) Q

C L

"O g D_ c o œ CD

o w C Q o TZ CL en C w CO a) CD O) R CJ (U oo O

OC £

CHAPTER 4

DISCUSSION

The present study describes the periodontal condition and the prevalence of

selected micro-organisms in an Indonesian population deprived of regular dental

care. The study sample consisted of young adolescents living in a remote village

on a tea estate, who, at the start of the investigation in 1987, were aged

between 15-25 years of age. As discussed in a previous paper describing cross-

sectionally the baseline data (1987), the prevalence of periodontal attachment

loss in this population can be considered high (Timmerman et al. 1998).

Van der Velden et al. (1996) evaluated the development of periodontitis in

spouses within this population and observed that at least 10 years of

cohabitation showed no influence on the periodontal condition in spouses.

Therefore, the role of a periodontally diseased subject as a source for the spread

of periodontitis was questioned. One possible microbiological cause of the high

prevalence of attachment loss was refuted by Van Winkelhoff et al. (1999),

who concluded that transmission of P. gingivalis between spouses and siblings

as a possible etiological factor was highly unlikely. An explanation was more

likely to be found in a genetic background based on the results of Van der

Velden et al. (1993) who investigated the effect of sibling relationships on the

periodontal condition in this study population. Loss of attachment was found to

be associated with sibship.

In the cross-sectional evaluation of the baseline data no significant association

between the clinical periodontal parameters and the studied micro-organisms at

a patient level was found (Timmerman et al 1998). However, at selected sites

with a pocket depth > 4 mm in conjunction with attachment loss > 4 mm, P.

gingivalis and spirochetes were more frequently detected compared to sites

without attachment loss in the same subjects. In the present study an attempt

was made to explain the association of local factors with local disease activity

in a constrained design using the Sampled Pocket as a patient level response

variable. Odds ratios as presented in Table 5 show that presence of P. gingivalis

and motile micro-organisms, predicts a 3x increased odds for disease

progression. In the logistic regression model, significant associations for disease

progression were found with the presence of motile micro-organisms and the

plaque score. Findings in the 2nd level of the regression analysis showed that

for the single site response model, the subsidiary factor to plaque was the

92

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

presence of P. intermedia. This is in agreement with studies that suggested that

the presence of P. intermedia is related to the level of oral hygiene (Wolfe et al.

1988, Lie et al. 1998).

The logistic regression analysis on the full mouth data showed that the main

background factor associated with progression of disease was age. Longitudinal

epidemiological studies have long since recognized that the prevalence of

periodontitis increases with age (Löe et al. 1986). Even in the limited age range

of this population ( 1 5 - 2 5 yrs) the odds ratio measuring the increase of the

chance for disease progression, as derived from the logistic regression analysis

was 1.15, per year of age. This results for example in an increased odds ratio of

disease progression by a factor of 4 ( = 1.1510) for subjects of 25 years of age

as compared to subjects of 15 years of age at baseline. This finding is

agreement with a recent publication by Suda et al. (1999), who found that the

amount of attachment lost at the end of a 3 year period showed a positive

correlation with age at baseline. They are also in corroboration with findings as

documented in tea workers in Sri Lanka, where the annual rate of attachment

loss increased after 27 years of age (Löe et al. 1978). Besides the effects of

ageing itself, another aspect, which is closely related to age is that a subject is

exposed to the bacterial attack for a longer period of time. This increases the

chance of development and progression of periodontitis. However, the logistic

regression analysis does not corroborate this mechanism since age was

identified as a main factor, although all microbiologic parameters were entered

into the analysis. Since age is an unmodifiable factor it should be considered a

'determinant' of disease progression (Genco 1996).

In the present study population with an age range of 1 5-25 at baseline, a mean

progression of 0.06 mm attachment loss per year was found in a time-span of 7

years (in total 0.07 mm per year). Compared to the Sri Lankan study (Löe et al.

1978) this appears to be a low rate. They found 0.29 mm in subjects with

moderately progressive disease and 1.04 mm in those with rapidly progressive

disease. The results of the present study are more in concordance with recently

published data on a Swedish population, in which a (radiological) progression

rate of 0.06 over a 17-year period is reported (Norderyd et al. 1999).

The number of surfaces with subgingival calculus at the baseline assessment

was found to be a main clinical factor associated with progression of disease in

the first level of the logistic regression analysis. Compared to those without

93

CHAPTER 4

subgingival calculus, subjects with calculus on all 12 surfaces of the Ramfjord

teeth had an approximately 9 times (1.212) higher odds for disease progression

during the 7 years time-span. Van Palenstein Helderman et al. (1998) studied

the relationship between the presence of calculus and recession associated

attachment loss in a Tanzanian population without regular dental care. High

correlation coefficients were found especially in the young age group of 20-34

years. They concluded that long-standing calculus is important with regard to

the onset of recession associated attachment loss. In untreated populations,

calculus is allowed to develop undisturbed. In a western population, where

calculus is removed, this parameter may not be a reliable 'risk marker'.

Pocket depth did not appear to be a risk marker, although this parameter is the

most important measurement in everyday clinical periodontal practice. The

present study showed no increase in pocket depth, whereas the attachment

loss did increase. The only way to explain this phenomenon is a decrease of the

distance between the gingival margin and the cemento-enamel junction. A

similar type of periodontal disease progression was observed in a study by

Yoneyama et al. (1988), who assessed in a cross sectional study of a Japanese

population, that attachment loss was accompanied by recession of the gingival

margin.

The main microbiological factor in the full mouth logistic regression analysis

associated w i th progression of disease was presence of A.

actinomycetemcomitans in the subgingival microbiota of the deepest bleeding

pocket without attachment loss at the time of the baseline measurements. No

association could be observed between disease progression and the presence of

micro-organisms on the mucous membranes and in the saliva. All associations

were found with the subgingival presence of the selected micro-organisms. This

corroborates the findings of Danser et al. (1996) who observed that periodontal

therapy influences the presence of putative periodontal pathogens in the

subgingival area, but not on the mucous membranes. In their study, the

reduction of the pathogens as a result of treatment did not result in any

concomitant change of their prevalence on the mucous membranes. The present

data thereby questions the influence of the microbiota that colonizes the

mucous membranes on the risk of the progression of periodontal disease.

The odds ratios as presented in Table 4 indicate that the presence of P.

gingivalis and motile micro-organisms is associated with an elevation ( ± 2x) of

94

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

the odds for disease progression. The relatively high prevalence of P. gingivalis

and motile micro-organisms in the NPDS group can explain the fact, that only A.

actinomycetemcomitans remains as a significant factor in the overall logistic

regression model.

Studies that have evaluated the subgingival presence of A.

actinomycetemcomitans, P. gingivalis, P. intermedia in relation to the

progression of treated periodontal disease also have concluded that the absence

or levels below certain thresholds of A. actinomycetemcomitans, P. gingivalis,

and P. intermedia, were the best guarantee for no further loss of attachment

(Wennström et al. 1987, Listgarten 1987, Slots et al. 1986, Dahlén et al. 1996,

Bragd et al. 1987, Haffajee 1997a, 1997b).

This longitudinal study assessed in a young population the relationship between

progression of the disease over a 7-year period and baseline clinical and

microbiological data. Close associations between several risk markers and

disease were established, but this survey cannot decide on cause and effect

relationships. Studies of such causal relationships require more precision than

population studies can offer. However, longitudinal studies are generally

considered to offer high quality evidence that an observed risk marker is indeed

a true risk factor (Genco 1996, Papapanou 1993). The results of this study

identified 3 main risk markers for disease progression at the full mouth level:

age, amount of subgingival calculus and subgingival presence of A.

actinomycetemcomitans.

ACKNOWLEDGEMENTS

The authors gratefully thank Mrs N. Dellemijn-Kippuw (Department of Oral

Microbiology, ACTA, Amsterdam) for technical assistance, ir. A.A.M. Hart

(Department of Clinical Epidemiology and Biostatistics, Faculty of Medicine,

University of Amsterdam) for advice and support on statistical analysis and

preparation of the manuscript and Mr. G.N. Wolffe for his aid during preparation

of the manuscript. In addition, the management of the Tea Company PTP XIII

and the medical staff of the Hospital Passir Junghun are greatly acknowledged

for their help and support in the execution of the research.

95

CHAPTER 4

REFERENCES

Baelum, V. , Luan, W . M . , Chen, X. & Fejerskov, 0 . (1997) A 10-year study of the progression

of destruct ive periodontal disease in adult elderly Chinese. Journal of Periodontology, 68 , 1033-

1042.

Björby, A. & Löe, H. (1967) The relative significance of different local factors in the init iation

and development of periodontal inf lammation. Journal of Periodontal Research 2 , Abstract 76-

77 .

Bragd, J . , Wikst röm, M., Dahlén, G. & Slots, J . (1987) The capability of Actinobacillus

actinomycetemcomitans, Bacteroides gingivalls and Bacteroides intermedius to indicate

progressive periodontit is; a retrospective study. Journal of Clinical Periodontology 14, 95 -99 .

Danser, M .M. , T immerman, M.F., Van Winkelhoff A . J . , De Graaff, J . & Van der Velden, U.

(1996) The effect of periodontal treatment on periodontal pathogens on the oral mucous

membranes. Journal of Periodontology, 67 , 478 -485 .

Dahlén, G., Wiks t röm, M. & Renvert S. (1996) Treatment of periodontal disease based on

microbiological diagnosis. A 5-year fol low-up on individual patterns. Journal of Periodontology,

67 , 879 -887 .

Davies, G.N. (1956) Dental conditions of the Polynesians of Pukapuka (Danger Island) Journal

of Dental Research 5, 7 3 4 - 7 4 1 .

Genco, R.J. (1996) Current view of risk factors for periodontal diseases. Journal of

Periodontology, 67 , 1041-1049 .

Grossi, S.G., Zambon, J .J . , Ho, A .W. , Koch, G., Dunford R.G., Machtei, E.E., Norderyd, P.M. &

Genco R.J. (1994) Assessment of risk for periodontal disease. I. Risk indicators for at tachment

loss. Journal of Periodontology, 65 , 260-267 .

Grossi, S.G., Genco R.J., Machtei, E.E., Ho, A .W. , Koch, G., Dunford R.G., Zambon, J .J . &

Hausmann, E. (1995) Assessment of risk for periodontal disease. II. Risk indicators for alveolar

bone loss. Journal of Periodontology, 66 , 23-29.

Haffajee, A .D . , Socransky, S.S. & Goodson, J .M . (1983) Clinical parameters as predictors of

destruct ive disease activity. Journal of Clinical Periodontology 10, 240 -257 -265 .

9 6

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

Haffajee, A .D . & Socransky, S.S. (1986) Attachment level changes in destructive periodontal

diseases. Journal of Clinical Periodontology, 13, 461-475 .

Haffajee, A .D . , Cugini, M.A., Dibart, S., Smith, C , Kent Jr., R.L. & Socransky, S.S. (1997a)

The effect of SRP on the clinical and microbiological parameters of periodontal diseases. Journal

of Clinical Periodontology, 24 , 324-334.

Haffajee, A .D . , Cugini, M.A., Dibart, S „ Smith, C , Kent Jr., R.L. & Socransky, S.S. (1997b)

Clinical and microbiological features of subjects who responded poorly to scaling and root

planing. Journal of Clinical Periodontology, 24 , 767-776.

Ladavalya, M.R.N. & Harris, R. (1959) A study of the gingival and periodontal condit ions of a

group of people in Chieng Mai Province. Journal of Periodontology, 30 , 219-222 .

Lie, M.A., Van der Weijden, G.A., Timmerman, M.F., Loos, B.G., Van Steenbergen, T .J .M. &

Van der Velden, U. (1998) Oral microbiota in smokers and non-smokers in natural and

experimentally induced gingivit is. Journal of Clinical Periodontology 25 , 677-686 .

Lindhe, J . , Haffajee, A .D. & Socransky, S.S. (1983) Progression of periodontal disease in adult

subjects in the absence of periodontal therapy. Journal of Clinical Periodontology 10, 4 3 3 - 4 4 2 .

Listgarten, M.A. (1987) A rationale for monitoring the periodontal microbiota after periodontal

t reatment. Journal of Periodontology 59, 439-444 .

Löe, H., Anerud, A. , Boysen, H. & Smith, M. (1978) The natural history of periodontal disease

in man. Study design and baseline data. Journal of Periodontology 49 , 607-620 .

Löe, H., Anerud, A. , Boysen, H. & Morrison E. (1986) Natural history of periodontal disease in

man. Rapid, moderate and no loss of attachment in Sri Lankan laborers 1 4 to 46 years. Journal

of Clinical Periodontology 13, 431 -440 .

Löe, H. & Morr ison, E. (1990) Epidemiology of periodontal of disease. In Contemporary

periodontics, eds. Genco, R.J., Goldman, H.M. & Cohen, D.W., 1st edit ion, pp.106-1 16, St.

Louis: Mosby.

Lövdal, A. , Arno, A. & Waerhaug J . (1958) Incidence of clinical manifestations of periodontal

disease in the light of oral hygiene and calculus formation. Journal of the American Dental

Association, 56 , 21-33.

97

CHAPTER 4

Mobley, E. & Smith, S.H. (1963) Some social and economic factors relating to periodontal

disease among young negroes: N o . 1 . Journal of the American Dental Association, 66 , 4 8 6 - 4 9 1 .

Norderyd, 0 . , Hugoson, A. & Grusovin, G. (1999) Risk of severe periodontal disease in a

Swedish adult populat ion. A longitudinal study. Journal of Clinical Periodontology 26 , 608 -615 .

Olsson, B. (1978) Periodontal disease and oral hygiene in Arussi province, Ethiopia. Community

Dental and Oral Epidemiology 6, 1 39-145 .

Papapanou, P.N. (1993) Epidemiology and natural history of periodontal disease. In Proceedings

of the 1st European Workshop on Periodontology, Session I, pp. 2 3 - 4 1 , Quintessence Books.

Polsky, V .W. (1996) Epidemiology of gingival and periodontal disease. In Clinical

Periodontology, eds. Caranza. F. A. & Newman, M. G., pp. 6 1 - 8 1 . Philadelphia: Saunders.

Ramfjord, S.P. (1959) Indices for prevalence and incidence of periodontal disease. Journal of

Periodontology 3 0 , 51 -59 .

Russell, A .L . & Ayers, P. (1960) Periodontal disease and socio-economic status in Birmingham,

Ala. American Journal of Public Health 50 , 206-209 .

Schei, O., Waerhaug, J . , Lövdal, A. & Arno, A. (1959) Alveolar bone loss as related to oral

hygiene and age. Journal of Periodontology, 30 , 7-16.

Silness, J . & Löe, H. (1964) Periodontal disease in pregnancy. II. Correlation between oral

hygiene and periodontal condit ion. Acta Odonto/ogica Scandinavica 22 , 121-135.

Slots, J . , Bragd, L., Wikst röm, M. & Dahlén G. (1986) The occurrence of Actinobacillus

actinomycetemcomitans, Bacteroides gingiva/is and Bacteroides intermedins in destruct ive

periodontal disease in adults. Journal of Clinical Periodontology 13, 570 -577 .

Suda, R., Cao, C F . , Suzuki, M., Hasegawa, K. & Sasa, R. (1999) At tachment loss in rural

Chinese children over a 3-year period. Community Dentistry and Oral Epidemiology, 27 , 216-

220 .

Timmerman, M.F., Van der Weijden, G.A., Abbas, F., Armand, S., Winkel, E.G., Van

Winkelhoff , A . J . & Van der Velden, U. (1998) Untreated periodontal disease in Indonesian

adolescents. Clinical and microbiological baseline data. Journal of Clinical Periodontology 25 ,

2 1 5 - 2 2 4 .

98

LONGITUDINAL DATA AND PROSPECTIVE RISK ASSESSMENT

Van Palenstein Helderman, W.H. , Lembariti, B.S., Van der Weijden, G.A. & Van 't Hof, M.

(1998) Gingival recession and its association wi th calculus in subjects deprived of prophylactic

dental care. Journal of Clinical Periodontology, 25, 1 06-1 1 1 .

Van der Velden, U. (1979) Probing force and the relationship of the probe tip to the periodontal

t issues. Journal of Clinical Periodontology 6, 106-1 14.

Van der Velden, U., Abbas, F., Armand, S., De Graaff, J . , Timmerman, M.F., Van der Weijden,

G.A., Van Winkelhoff , A . J . & Winkel, E.G. (1993) The effect of sibling relationship on the

periodontal condit ion. Journal of Clinical Periodontology 20 , 683-690 .

Van der Velden, U., Van Winkelhoff, A . J . , Abbas, F., Arief, E.M., Timmerman, M.F., Van der

Weijden, G.A. & Winkel, E.G. (1996) Longitudinal evaluation of development of periodontal

destruct ion in spouses. Journal of Clinical Periodontology 23, 1014-1019.

Van Winkelhoff , A . J . , Laine, M.L., Timmerman, M.F., Van der Weijden, G.A., Abbas, F.,

Winkel , E.G., Arief, E.M. & Van der Velden, U. (1999) Prevalence and serotyping of

Porphyromonas gingivalis in an Indonesian population. Journal of Clinical Periodontology 26 ,

301 -305 .

Wennst röm, J.C. , Dahlén, G., Svensson, J. & Nyman, S. (1987) Actinobacillus

actinomycetemcomitans, Bacteroides gingivalis, Bacteroides intermedius: Predictors of

at tachment loss? Oral Microbiology and Immunology 2, 158-163.

Wol fe, L.F., Liljemark, W.F., Philstrom, B., Schaffer, E.M., Aeppli , D.M., & Brandt, C L . (1988)

Dark-pigmented Bacteroides species in subgingival plaque in adult patients on a rigorous recall

program. Journal of Periodontal Research 23, 170-174.

Yoneyama, T., Okamoto, H., Lindhe, J . , Socransky, S.S. & Haffajee, A.D. (1988) Probing

depth, at tachment loss and gingival recession. Findings from a clinical examination in Ushiku,

Japan. Journal of Clinical Periodontology 15, 5 8 1 - 5 9 1 .

99