PERIODONTAL CONDITIONS IN SWEDISH ADOLESCENTS

Carina Källestål

Umeå 1991

PERIODONTAL CONDITIONS IN SWEDISH ADOLESCENTS

AKADEMISK AVHANDLING

som med vederbörligt tillstånd av Odontologiska fakulteten vid Umeå universitet för avläggande av odontologie doktorsexamen kommer att offentligen försvaras i föreläsningssal B, Byggnad ID, Odontologiska

kliniken, Umeå, fredagen den 18 oktober 1991, kl 09.00

av

Carina Källestål

Avhandlingen baseras på följande delarbeten:

I Källestål C, Matsson L. Criteria for assessment of interproximalbone loss on bite-wing radiographs in adolescents. J Clin Perio- dontol 1989;16:300-4.

II Källestål C, Matsson L. Marginal bone loss in 16-year-oldSwedish adolescents in 1975 and 1988. Accepted for publication inJ Clin Periodontol 1991; 18.

III Källestål C, Matsson L, Holm A-K. Periodontal conditions in agroup of Swedish adolescents. (I) A descriptive epidemiologic study. J Clin Periodontol 1990;17:601-8.

IV Källestål C, Matsson L. Periodontal conditions in a group ofSwedish adolescents. (II) Analysis of data. J Clin Periodontol1990;17:609-12.

V Källestål C, Matsson L, Persson S. Proximal attachment loss inSwedish adolescents. Accepted for publication in J Clin Perio­dontol 1991;18.

VI Källestål C, Uhlin S. Buccal attachment loss in Swedish adoles­cents. Accepted for publication in J Clin Periodontol no. 7, 1992.

Umeå 1991

ABSTRACTKällestål Carina. 1991. Periodontal conditions in Swedish adolescents. Umeå University Odontological Dissertations Abstract No. 41, ISSN 0345-7532, ISBN 91- 7174-603-X.

This thesis describes the periodontal conditions in adolescents who have taken part in an organized dental health program since pre-school age. Loss of tooth support was estimated from bone loss measurements and measurements of probing attachment level. To establish a threshold for bone loss, a methodological study was performed on the normal appearance of the proximal alveolar bone. In a subsequent study, bone loss was used to compare the periodontal conditions in two cohorts of adolescents (n=400) who were 16 years old in 1975 and 1988, respectively. In a more comprehensive cross- sectional study, the periodontal conditions of 287 16-year-olds and 283 18-year-olds from northern Sweden were described. Information on periodontal conditions as well as on background factors thought to be connected with periodontal disease was gathered by means of a clinical and radiographic examination and an interview. In this study, two groups were distinguished, one with attachment loss at buccal surfaces and the other with proximal attachment loss. In a following analysis the groups with attachment loss at different surfaces were compared for differences in background factors. Each of the groups was then examined separately in two case-referent studies.

The main findings of the studies were:► In adolescents without any sign of gingival inflammation or loss of probing attach­ment the distance between the cemento-enamel junction and the alveolar crest was in the 0-2 mm range; thus, for epidemiologic purposes a distance of >2 mm between the alveolar crest and the cemento-enamel junction seems appropriate as a criterion of bone loss in adolescents.► The same prevalence of bone loss in 16-year-olds (3.5%) was found in 1975 and 1988.► In the cross-sectional study the prevalence of probing attachment loss was 9% in 16- year-olds and 19% in 18-year-olds. The compromised attachment was found mainly at buccal surfaces. Proximal loss of attachment was found in 3% and 5% of the individu­als, respectively.► The group with proximal attachment loss had more plaque, bleeding, calculus and pocket depths ^ 4 mm than the group with buccal loss. In the latter group, toothbrush- ing was more frequent than in the group without attachment loss. The results indicate different etiologies of loss of tooth support at different tooth-surfaces at these ages.► Various factors may contribute to the loss of proximal attachment in young individu­als. A majority of those with attachment loss had solitary affected sites, some without signs of inflammation, possibly indicating that the destruction of the periodontium had occurred earlier. Two cases were tentatively given the diagnosis periodontitis, thus giv­ing a prevalence of 0.3% in this population.► Factors related to the anatomy of the alveolar process seem to be associated with buc­cal attachment loss.

Keywords: epidemiology; oral; periodontal conditions; periodontitis; alveolar bone loss; attachment loss; gingival recession.

C Källestål, Department of Pedodontics, University of Umeå, S-901 87 Umeå, Sweden.

UMEÀ UNIVERSITY ODONTOLOGICAL DISSERTATIONS Abstract No. 41 - ISSN 0345-7532

From the Department of Pedodontics, University of Umeå, Sweden

PERIODONTAL CONDITIONS IN SWEDISH ADOLESCENTS

Carina Källestål

Umeå 1991

Copyright 1991 by Carina Källestål ISBN 91-7174-603-X

Printing and binding by Nyheternas Tryckeri KB, Umeå, 1991

CONTENTS

Abstract ......................................................................................... 5Preface .......................................................................................... 6Introduction .................................................................................. 7Periodontal health and disease in adolescents ..............................13Measuring periodontal health and disease ................................... 17Own investigations and findings ...................................................20

Paper I ....................................................................................20Paper II .................................................................................. 21Paper 111 ................................................................................. 22Paper IV .................................................................. 24Paper V .................................................................................. 25Paper VI ................................................................................. 28

Discussion ...................................................................................... 29Conclusions ....................................................................................39Acknowledgements ......................................................................... 40References ...................................................................................... 43Appendix: Papers I-VI ...................................................................51

ABSTRACTKällestål Carina. 1991. Periodontal conditions in Swedish adolescents. Umeå Univer­sity Odontological Dissertations Abstract No. 41, ISSN 0345-7532, ISBN 91-7174-603- X.

This thesis describes the periodontal conditions in adolescents who have taken part in an organized dental health program since pre-school age. Loss of tooth support was estimated from bone loss measurements and measurements of probing attachment level. To establish a threshold for bone loss, a methodological study was performed on the normal appearance of the proximal alveolar bone. In a subsequent study, bone loss was used to compare the periodontal conditions in two cohorts of adolescents (n=400) who were 16 years old in 1975 and 1988, respectively. In a more comprehensive cross- sectional study, the periodontal conditions of 287 16-year-olds and 283 18-year-olds from northern Sweden were described. Information on periodontal conditions as well as on background factors thought to be connected with periodontal disease was gathered by means of a clinical and radiographic examination and an interview. In this study, two groups were distinguished, one with attachment loss at buccal surfaces and the other with proximal attachment loss. In a following analysis the groups with attachment loss at different surfaces were compared for differences in background factors. Each of the groups was then examined separately in two case-referent studies.

The main findings of the studies were:► In adolescents without any sign of gingival inflammation or loss of probing attach­ment the distance between the cemento-enamel junction and the alveolar crest was in the 0-2 mm range, thus; for epidemiologic purposes a distance of >2 mm between the alveolar crest and the cemento-enamel junction seems appropriate as a criterion of bone loss in adolescents.► The same prevalence of bone loss in 16-year-olds (3.5%) was found in 1975 and 1988.► In the cross-sectional study the prevalence of probing attachment loss was 9% in 16- year-olds and 19% in 18-year-olds. The compromised attachment was found mainly at buccal surfaces. Proximal loss of attachment was found in 3% and 5% of the individu­als, respectively.► The group with proximal attachment loss had more plaque, bleeding, calculus and pocket depths >4 mm than the group with buccal loss. In the latter group, toothbrush- ing was more frequent than in the group without attachment loss. The results indicate different etiologies of loss of tooth support at different tooth-surfaces at these ages.► Various factors may contribute to the loss of proximal attachment in young individu­als. A majority of those with attachment loss had solitary affected sites, some without signs of inflammation, possibly indicating that the destruction of the periodontium had occurred earlier. Two cases were tentatively given the diagnosis periodontitis, thus giv­ing a prevalence of 0.3% in this population.► Factors related to the anatomy of the alveolar process seem to be associated with buc­cal attachment loss.

Keywords: epidemiology; oral; periodontal conditions; periodontitis; alveolar bone loss; attachment loss; gingival recession.

C Källestål, Department of Pedodontics, University of Umeå, S-901 87 Umeå, Sweden.

5

PREFACE

This thesis is based on the following papers, which are referred to in thetext by their Roman numerals:

I Källestål C, Matsson L. Criteria for assessment of interproximalbone loss on bite-wing radiographs in adolescents. J Clin Perio- dontol 1989;16:300-4.

II Källestål C, Matsson L. Marginal bone loss in 16-year-oldSwedish adolescents in 1975 and 1988. Accepted for publication inJ Clin Periodontol 1991;18.

III Källestål C, Matsson L, Holm A-K. Periodontal conditions in agroup of Swedish adolescents. (I) A descriptive epidemiologic study. J Clin Periodontol 1990;17:601-8.

IV Källestål C, Matsson L. Periodontal conditions in a group ofSwedish adolescents. (II) Analysis of data. J Clin Periodontol1990;17:609-12.

V Källestål C, Matsson L, Persson S. Proximal attachment loss inSwedish adolescents. Accepted for publication in J Clin Perio­dontol 1991 ; 18.

VI Källestål C, Uhlin S. Buccal attachment loss in Swedish adoles­cents. Accepted for publication in J Clin Periodontol no. 7, 1992.

6

INTRODUCTION

The concept o f periodontal health and disease

The tissues supporting the teeth - the periodontium - consist of the gingiva, periodontal ligament, root cementum and alveolar bone. Clinic­ally, a healthy periodontium is characterized by a gingiva which has a pale pink color and is firm and non-mobile, shows a scalloped outline of the soft tissue margin, occupies the entire space below the contact area of neighboring teeth and does not bleed on gentle probing in the gingi­val sulcus. Accumulation of bacteria in the dento-gingival area results in gingival inflammation, gingivitis. The clinical signs of gingivitis - red­ness, gingival bleeding and edema - appear within 1-3 weeks after suspension of all oral hygiene procedures (Löe et al. 1965). Gingivitis, reversible if oral hygiene measures are reinstituted, may develop into a lesion characterized by destruction of connective tissue attachment to the root surface and alveolar bone loss, i.e. periodontitis.

Many of the earlier epidemiological studies of gingivitis and periodonti­tis, these conditions generally seen as a disease entity, reported a reciprocal relationship between gingivitis and periodontitis, the former being high and the latter low in young individuals and the reverse in older groups. This led to the inference that in the absence of treatment, gingivitis will automatically progress to periodontitis. Moreover, periodontitis was generally seen as a slowly progressing chronic disease which destroyed the natural dentition and caused tooth loss, and thus also had severe effects on general health (WHO 1961).

In the late 1970s and early 1980s this conception of the periodontal diseases began to be revised mainly due to the development of microbiological methods for cultivating anaerobic bacteria, and a more analytical approach in epidemiology. It was shown that the relative proportions of bacteria differed in samples from subjects with clinically healthy gingiva, gingivitis and periodontitis, respectively (Slots 1977a,b, Slots et al. 1978, Listgarten & Helldén 1978) and that some

7

species were closely associated with specific disease patterns (Slots 1976, Newman & Socransky 1977). More careful examination of epidemiological data and reports from the developing countries showed that usually only a limited proportion of the studied populations had periodontitis and that this proportion seemed to be similar in popula­tions with a wide range of access to prevention and treatment. Obvious­ly gingivitis did not always progress to periodontitis and even in the ab­sence of treatment, only a fraction of the sites with gingivitis later de­velops periodontitis (Löe et al. 1986). Moreover, progression of the disease did not seem to be linear (Haffajee et al. 1983, Lindhe et al.1983), which had been inferred earlier from analyses of data using pooled mean values. The pattern of progression that emerged from analyses of individual sites seemed to be more episodic, not related to the earlier disease level or age and characterized by acute "bursts" with tissue destruction, followed by long periods of remission (Goodson et al. 1982, Socransky et al. 1984, Listgarten 1986). This so called burst theory for the pattern of progression of periodontitis has, however, been questioned by some authors (Imrey 1986, Ralls & Cohen 1986).

Diagnosis and prediction o f disease activity

As the transition from a reversible inflammation restricted to the gingi­val tissues to an involvement of deeper periodontal tissues is not fully understood, no "gold standard" of disease has been identified. Instead, it is the perception of the disease, historically dependent on current re­search and clinical practice, that has influenced the means of measure­ment. Therefore, as periodontal disease in the 1950-60s was considered a single entity which began with gingivitis and progressed to periodonti­tis and tooth loss, the indices used for assessing disease represented a combination of these conditions (PI described by Russel 1956, PDI described by Ramfjord 1959).

When periodontitis was distinguished from gingivitis, the use of indices was abandoned; destroyed tissue attachment and clinical signs of inflammation were measured and presented separately. The amount of

destroyed tissue attachment is commonly measured by an indirect clinic­al technique, estimating by probing the distance between the connective tissue attachment and the cemento-enamel junction, while the degree of gingival inflammation often is studied by probing in the gingival crevice to assess bleeding. Assessment of alveolar bone loss, using radiographs, is another way of estimating lost periodontal tissue. Data on lost perio­dontal tissue are considered to represent past disease experience. The shortcomings of these assessments, which neither distinguish between attachment lost due to periodontitis and other causes nor evaluate the activity of the disease, are widely acknowledged (Kornman 1987, Griffiths et al. 1988, Listgarten 1988). To overcome the problem of not being able to measure active disease, repeated probing measurements to evaluate changes in attachment levels over time are now used in clinical research and practice (Badersten et al. 1985, Goodson 1986).

The identification, in most populations, of a relatively small proportion with periodontitis highlights the need for markers of active disease with which to indicate patients or sites at risk for additional periodontal breakdown. Suzuki (1988) outlined three possible hypotheses for the pathogenesis of periodontitis: a) direct tissue destruction caused by bacterial plaque and metabolic products, b) immune hyperresponsive­ness, and c) immune deficiencies (Fig. 1). These hypotheses are the scientific background to the search for biological markers of active breakdown of tissue or, even better, markers predicting this breakdown, as well as markers that might stand for characteristics of different forms of periodontitis. Biological markers have been suggested (Curtis et al. 1989, Wilton et al. 1988, 1989, Maiden et al. 1990) but are not yet at hand. Until markers of active disease have been found, measurements of probing attachment level, probing pocket depth, radiographic bone loss, gingival bleeding, plaque and calculus are what are available for the assessment of periodontal conditions.

9

Bacteria of Dental Plaque

^ i ^Activation of Direct Bacterial Immune

Immune Mechanisms Tissue Damage Dysfunctions

^ i /Periodontal Diseases

Fig. 1. Current concepts of pathogenesis of periodontitis according to Suzuki (1988).

Classifications of periodontal diseases have been suggested by several authors and likewise reflect the changing perception of periodontitis. Table 1 presents some of the most recent classifications. Disregarding diagnoses concerning gingivitis and diagnoses connected to some gener­al diseases, it can be said of the classifications that prepubertal, juvenile and post-juvenile periodontitis are described as having a fast rate of progression and a localized or generalized intra-oral distribution, while adult periodontitis is described as having a slow rate of progression and a generalized intra-oral distribution.

10

Table 1. Recent classifications of periodontal diseases: Suzuki (1988), American Associa­tion of Periodontology (AAP) (Caton 1989), Topic (1990).

Suzuki's classification

Forms of gingivitis1. Plaque-associated gingivitis2. ANUG3. Steroid hormone-influenced gingivitis4. Medication-influenced gingival overgrowth5. Other forms of gingivitis

Forms of periodontitis1. Adult periodontitis2. Rapidly progressive periodontitis: type A3. Rapidly progressive periodontitis: type B4. Juvenile periodontitis5. Postjuvenile periodontitis6. Prepubertal periodontitis

AAP's classificationI. Adult periodontitis

II. Early-onset periodontitisA. Prepubertal periodontitis

1. Generalized2. Localized

B. Juvenile periodontitis1. Generalized2. Localized

C. Rapidly progressive periodontitis

III. Periodontitis associated with systemic disease

IV. Necrotizing ulcerative periodontitis

V. Refractory periodontitis

Topic's classification

Acute conditions1. Gingivitis

(a) Specific- Ulceromembranous- Herpetic- Coccal(b)Non-specific

2. Periodontitis(a) Periodontal abcess(b)Periocoronitis

Chronic conditions1. Gingivitis

(a) Plaque associated(b) Symptomatic

2. Periodontitis(a) Simplex (adult)(b) Complex- Generalized forms

Rapidly progressive: type Atype B

Prepubertal

- Localized forms Juvenile Post-juvenile Prepubertal

(c) Symptomatic

3. Recession

11

Epidemiology o f periodontal disease

Due to the lack of diagnostic standards for periodontitis no unequivocal figure can be given for the prevalence of periodontitis among adults. Prevalence figures are highly dependent on how the condition is defin­ed. Despite these problems, inferences are frequently drawn from epidemiological studies. Reviewers of prevalence studies base their estimates on the reported measures of probing attachment loss >5-7 mm or probing pocket depth >6 mm and sometimes on pronounced bone loss. Several reviewers have concluded, from studies performed in geographically and ethnically different populations (Hugoson & Jordan 1982, Burt et al. 1985, Baelum et al. 1986, 1988, Miller et al. 1987, Yoneyama et al. 1988), that advanced periodontal breakdown affects only a small fraction of the populations. Ismail et al. (1990) considered that the world-wide prevalence of advanced loss of periodontal attachment is in the 5-30% range and Johnson et al. (1988) stated that the prevalence of severe destructive periodontitis varies between 7% and 15%.

However, there are also populations where the prevalence of periodon­titis is reportedly high. In a review by Pilot (1986) of research using the Community Periodontal Index of Treatment Need (CPITN) (Ainamo et al. 1982), reported to WHO, studies from five countries detected pock­ets > 6 mm in 28-75% of the populations. Studies on the natural history of periodontitis involving populations in Sri Lanka and Norway (Löe et al. 1978) revealed that the loss of attachment among tea laborers in Sri Lanka was far greater than among Norwegians. Differences in the distribution of periodontitis have been discussed in the literature and suggested reasons for them are malnutrition and ethnicity.

Recent studies carried out among populations enjoying a good level of nutrition seem to support the statement that just a fraction is affected by severe periodontitis, measured as mentioned above. Studies performed in Scandinavia report figures between 1.5% and 23% (Hailing & Björn 1986, Sewón & Parvinen 1988, Papapanou et al. 1988, Hansen et al. 1990) and in other European countries 4.5%-20% (Vrbic et al. 1987,

12

Schürch et al. 1988, Flores-de-Jacoby et al. 1989, Skaleric & Kovac- Kavcic 1989). In North America, 11% is reported from Canada (Hoover & Tynan 1986), while Capilouto & Douglass (1988), analyzing the national examinations performed in the USA, concluded that approximately 20% of the US population is affected by periodontitis. It is debatable, however, whether a fifth or almost a quarter of a popula­tion should be considered a low fraction.

PERIODONTAL HEALTH AND DISEASE IN ADOLESCENTS

Juvenile periodontitis

The classifications suggested for periodontal diseases (Table 1) general­ly start from the age at onset of the disease. Localized juvenile periodontitis and generalized juvenile periodontitis seem to be the types of destructive periodontal disease that occur in adolescence. Localized juvenile periodontitis is characterized by a rapid loss of connective tiss­ue attachment and alveolar bone around the permanent first molars and incisors (Baer 1971). The extensive destruction of tissue is generally not considered commensurate with the amount of local irritants. Studies of this type of periodontitis have focused on the etiological relevance of specific microbes or combinations thereof (Newman et al. 1976, Slots 1976, Newman & Socransky 1977, Mandell 1984, Christersson et al. 1985, Asikainen 1986, Asikainen et al. 1987), defects in the host im­mune system (Lehner et al. 1974, Schenkein et al. 1976, Budtz- Jörgensen et al. 1978, Suzuki et al. 1984a,b, van Dyke et al. 1986), development disorders of tooth cementum (Lindskog & Blomlöf 1983, Blomlöf et al. 1986), and hereditary factors (Benjamin & Baer 1967, Fourel 1972, Jorgensen et al. 1975, Melnick et al. 1976, Saxén 1980b, Suzuki 1988). Of the microorganisms suggested as important in the pathogenesis of juvenile periodontitis, Actinobacillus actinomycetem-

13

comitans is considered of special interest (Zambon 1985). The preva­lence of juvenile periodontitis varies between studied populations but is often defined as rare, seldom affecting more than one percent of the po­pulation (Saxén 1980a, Kronauer et al. 1986, Saxby 1984, 1987).

Epidemiology o f periodontal disease in adolescents

A number of epidemiological studies of periodontal conditions in adole­scents have been performed since the 1950s. Recent studies are listed in Table 2. The results, presented by several authors as prevalence of periodontitis, cover a wide range. In studies using probing attachment loss as a measure of periodontitis, prevalences range from 3% to 89% (Mann et al. 1981, Wolfson & Lewis 1985, Clerehugh & Lennon 1986, Wolfe & Carlos 1987). In those using radiographically assessed bone loss, prevalences vary from 0.06% to 51% (Hull et al. 1975, Davies et al. 1978, Blankenstein et al. 1978, Macgregor 1980, Latcham et al. 1983, Gjermo et al. 1984). To some extent this variation can be explain­ed by differences in examination methods and criteria, but true varia­tions between populations can also exist.

In Scandinavia a few studies have been published. Bone loss was report­ed to be present in 11% of 15-year-olds and in 4.5% of 14-year-olds from Norway (Hansen et al. 1984, Aass et al. 1988), in 4% of 15-year- olds in Sweden (Hugoson et al. 1981) and in 2% and 3%, respectively, of Swedish 16- and 19-year-olds (Crossner & Unell 1986). In studies of Swedish army recruits, bone loss was found in 8% and 13% (Bäckman et al. 1981, Björn & Halling 1983). In Denmark, probing attachment loss was reported in 11% of 14-16-year-olds (Hoover et al. 1981a). The authors used different methods as well as different criteria and some samples are small, which makes it difficult to get a clear picture of the periodontal status in Scandinavian adolescents. This confusion is height­ened by the lack of discussion on the significance of the epidemiological findings in relation to the existing classifications of periodontal disease (Table 1).

14

Table

2. E

pidem

iolog

ical s

urvey

s of

perio

dont

al co

nditio

ns

in ad

olesc

ents.

AL

= pro

bing

attac

hmen

t lev

el; X

-ray

= ra

diogr

aphic

diag

nosis

of

bone

loss;

CEJ

= dis

tance

from

ceme

nto-

enam

el ju

nctio

n. Fo

r ref

eren

ces,

see R

eferen

ce se

ction

.

8 «jg « ^5 «I 2

£ 8 .

vo vo P 00 vo

T t. od

O (N

VOTt »-H

Tj- 00 CN rH

co oT t vn cm co os r *os" r i v i oo P *>. n *1 ‘00 ^

00 V)CMt-h cm

.2I’uu

£ CI 8CM CMAl Al

ao a a g o o E g g

vo £ £ h n tn

aao o£ £ £ £

a a a a a a au o o o o o u £ £ £ £ £ £ £ £ £ £ £ £ £ £

s s £

'< '< ><

H ( M C M ^ C M n ( M n N ( M C MAl J h ^ Ala s s s s j j s ? ?

ao££

cä* 03* c3*

£ £ £ £

CM ^ ^Ss Al A 2

au£ £ £ £ £ £ £ £

CM T -i CM (NAl Al Al

U m

â 'S*« 33 H-i'S *.3 «M .► O T3S ä

o »® ä

Z a

c<3£§

Ö

T3 X) C C ^•o ,2 öo c oo c ,2 £ WÖJQ W -

w £ 2

c<3£G<D

S-a_, Ü1= SP .2 Ztwo „

<D

1CO

ss 5i ’S

(DIO

'5CO

cö

« iCO <L>P O

s3

Gs3

P•a

G S3

§ Ê * i,

§ § E 3«

TDGS3

< 3 2

2 3 vo vo<«/

z»o

>»Tt

>1CM

CO CMCM vo

vt3Z

1CM

1T*

vo"Tt

oos

cm"s

COr - p 3 8

c i00

co"00

vnCM

cn CO CM CM CO CO

CM Ti­ VO 00 00 O t-Hr ­ r - r* r - 00 00 00 00 00

os os os Os os os os os Os os

i—*S >n g v1» 'l-S^ CM S3

a?! z* ̂ SfCM CM 'O

•o rj G o c

r3

ÏS

i

Ss

i

Gs3O0

1

*2W)G

wcCO

Ss

ÎJ3

Z <CO

COp

N2

PQ

ofc

oz Ss

soCOSS

Ssso

DÎS

oz SSr -

DSs

Ssr -

iSS Ss 1 os v-H Os Ss i vo Tt

VOvo vo CM

SS>VTi­ Tf

TfT—(

Tt 1CM

Os" CM oo" so r —H r-~ VOTt Tf

CM̂ vo"r - r-"

os"CM

oo" sor ^

SOVO r-"

o T—HCM cm" os CM s cm" Tt CO

Tj- Tt Tt vo SO SO r - 00 OS o00 00 00 00 00 00 00 00 00 os osos OS Os os OS os os os os OS Os

uq

Sc*CO<L>

5O -s

*< « J-J2 « g~£ sM ° ^ C

> 1 cS3 O

C 0 P

ü m -hy 4 « 03a S « «•S g < z g g» g I

M 0 0 o jl§ s g>åS S k S

cooc

13K<«P

§«I I

13t 15PJ PP

CP PQT3 o G2 B OCO<D

.« § o ac

s2CO

13

2CO

<S

I I

G 0 h O *■*£ W3g 5P eö % O

•Skö S > 2 H *ao o

O £

O

c3 c+-> S3 Ü c

=1o Z

o ^> CÖ ° 5S «ä j) 5> CU CQ

15

The Scandinavian studies in which probing attachment loss or bone loss has been reported in 2-13% of adolescents concern populations with access to free, regular dental care, including preventive programs, from pre-school age. In these populations there has been a clearly document­ed decrease in caries prevalence (Bilie et al. 1986, Birkeland & Bragelien 1987, Vehkalahti et al. 1990, Källestål et al. 1990). In Sweden, where water fluoridation is not used and dietary habits do not seem to have changed significantly over the last decades (Persson 1984), this reduction of caries is commonly ascribed to increased information about dental health and to the free, organized dental care, based on annual check-ups from 3 to 19 years of age, leading to improved oral hygiene and an increased use of fluorides (Koch 1982). Whether the improved oral hygiene and the participation in the Public Dental Health Service have also resulted in an improvement in the periodontal health of adolescents is not known.

Gingival recessions in adolescents

The studies in Table 2 using probing attachment loss do not differentiate between attachment loss on different tooth surfaces. The radiographic studies of bone loss use posterior bite-wing radiographs and therefore include only posterior proximal surfaces. When attachment is lost to the extent that the root is exposed, or the coronal edge of the marginal gingiva is located at the cemento-enamel junction, the condition is commonly referred to as gingival recession (Gorman 1967, Paterson 1979). Gingival recession is mostly found on the buccal surfaces of teeth. In adolescents, the reported prevalence of gingival recession varies between 10 and 74% (Gorman 1967, O'Lleary et al. 1971, Björn et al. 1981, Mierau & Fiebig 1986, Ainamo et al. 1986, Paloheimo et al. 1987, Miller et al. 1987, Frentzen et al. 1989). High figures have been reported in some Scandinavian studies: 62% for 15-year-olds from Malmö, Sweden (Björn et al. 1981) and 74% for 17-year-olds from Espoo, Finland (Ainamo et al. 1986).

Several etiological factors behind gingival recession have been discussed

16

in the literature (for reviews see Woofter 1969, Paterson 1979, Watson1984). Studies of associations between recession and these factors have concentrated on three aspects: toothbrushing habits (Sangnes & Gjermo 1976, Björn et al. 1981, Ainamo et al. 1986, Mierau & Fiebig 1986, Paloheimo et al. 1987, Frentzen et al. 1989), malposition of teeth (Gorman 1967, Modéer & Odenrick 1980) and anatomy of the gingiva and the underlying bone, i.e. the width of the attached or keratinized gingiva (Tenenbaum 1982, Schoo & van der Velden 1985, Wennström1987) and dehiscences of alveolar bone (Löst 1984). An association has also been found between the use of smokeless tobacco and gingival recession (Offenbacher & Weathers 1985). The gingival recession is sometimes included in the measured attachment loss in epidemiological studies of periodontal conditions (Bay 1972, Hoover et al. 1981a, Ånerud et al. 1983). The validity of bringing buccal surfaces into stud­ies of periodontal disease in adolescents is questioned by Clerehugh et al. (1988). Neither the relation between the attachment loss found on buccal and proximal surfaces, respectively, nor the relative importance of all the factors suggested as etiological for gingival recession seem to have been studied.

MEASURING PERIODONTAL HEALTH AND DISEASE

The problem with the indices for assessment of periodontal diseases, i.e., the mix of different variables such as bleeding and pocket depth or attachment level, also appplies to the world-wide Community Perio­dontal Index for Treatment Needs (CPITN) (Ainamo et al. 1982). More­over, this index is basically constructed to evaluate treatment needs, not disease prevalence. The Extent and Severity Index proposed by Carlos et al. (1986) will provide information on the disease in a population but was not constructed to estimate the prevalence of disease. Thus, these indices are less suitable for examinations of prevalence of periodontal

17

conditions.

The clinical parameters recorded during a periodontal examination in­clude measurements based upon the signs and symptoms of periodontal diseases as well as factors thought to predispose to periodontal destruc­tion. The parameters principally assess the level of oral hygiene, the degree of gingival inflammation and the amount of lost connective tissue support.

Oral hygiene is usually assessed from measurements of plaque and calculus. Poor oral hygiene has generally been regarded as a risk factor for periodontal diseases, alhough some authors have argued recently that it may be of minor importance (Johnson et al. 1988). There are different ways of assessing plaque and numerous indices are available but there seems to be a concensus about the usefulness of all methods for most purposes (Marthaler 1986, Fischman 1986, Johnson et al.1988). An important feature of plaque assessments is that they must be viewed as transient or cross-sectional.

Gingival bleeding upon stimulation is widely accepted as a clinical sign of gingival inflammation (Poison & Caton 1985). However, some authors have pointed out that probing evaluates only the conditions of superficial tissues, or pocket wall epithelium, and may not relate to the condition of the deeper gingival connective tissue and the periodontal ligament (Griffiths et al. 1988).

Probing measurements, i.e. assessments of probing attachment level and probing depth, have been discussed for validity, reproducibility and interpretation (Griffiths et al. 1988). The common clinical method to assess attachment level is an indirect measurement technique described by Ramfjord (1959) where firstly, the distance from the gingival margin to the base of the pocket is measured (pocket depth), and secondly, the cemento-enamel junction is located and the distance from this to the gingival margin is recorded. The difference between the two distances gives an indirect measure that has several names, among which are: loss of periodontal attachment, probing attachment level, clinical attachment

18

level. In the papers on which this thesis is based, the attachment level is estimated by probing as described above; when a level chosen as a crite­rion is detected, atttachment loss is considered present and is designated by the abbreviation AL. In the following text, probing attachment level stands for the measurement technique and attachment loss for the pre­sence of deficient attachment as defined by a criterion.

The validity of these measurements based on probing has been question­ed (Listgarten 1980). The depth of probe penetration is highly depen­dent on the inflammatory status of the gingival tissues. Other factors affecting the degree of penetration into the underlying tissues include the diameter of the probe tip and the force applied during probing. The reproducibility of probing measurements has been investigated and the results show that the majority of reexaminations are in the range of ±1.5 mm (Badersten et al. 1984).

The major shortcomings of the probing method are the relatively low reproducibility and the fact that single measurements do not differentiate between present and past disease activity. Although these shortcomings are acknowledged, measurement of probing attachment level is considered to be the primary way to assess the presence of destructive periodontal disease.

Radiographic estimates o f bone loss give a history of hard tissue changes. The criteria for assessment of bone loss vary. Some authors have used the criteria described by Hull et al. (1975), i.e., crestal irregularity, widening of the marginal periodontal ligament space, and a distance of more than 3 mm from the cemento-enamel junction to the alveolar crest. Others have used these criteria but have changed the dis­tance cemento-enamel junction to alveolar crest to 1 or 2 mm. Others have used the criterion described by Jorkjend & Birkeland (1976), a distance cemento-enamel junction to alveolar crest >2 mm, with no additional subjective evaluation of the appearance of the marginal bone. Although there have been discussions about whether the normal appear­ance of the alveolar crest differs at different ages, due to developmental changes and physiological ageing (Boyle et al. 1973), the above criteria

19

are often used irrespective o f age.

Information on the normal appearance of the alveolar crest is scarce and the criteria used to estimate bone loss are not as well scrutinized as those for probing attachment level.

OWN INVESTIGATIONS AND FINDINGS

Paper I

Adequate criteria for bone loss in adolescents are lacking and presup­pose knowledge of the normal appearance of the alveolar crest in young and healthy individuals. Therefore a study was performed to determine the normal radiographic distance between the cemento-enamel junction and the alveolar crest in adolescents.

In the Public Dental Health Service, in which 97% of the adolescents in the northern region take part, the gingival status (% bleeding units) of each individual at the age of 16 years is reported to the dental health authorities. The bleeding score is kept in the subject's record. For this study, of a total patient pool of about 600, 30 18-year-olds with a record of a clinically healthy gingiva at the age of 16 were chosen.

In a clinical examination, all mesial and distal sites from the first pre­molars to the mesial surface of the second molars were examined for bleeding on probing, probing attachment level, defective fillings, supra- and subgingival calculus. Four posterior bite-wing radiographs were taken. All sites with bleeding, probing attachment loss, defective fillings or calculus were excluded from the following radiographic analysis. In total 737 sites were included. The radiographs were evaluated in a dark room, using a standard light box with adjustable light intensity. A pair of observation binoculars was used, equipped with a scale allowing measurements to the nearest 0.5 mm. All proximal surfaces from the

20

mesial surface of the first premolar to the mesial surface of the second molar in every quadrant were evaluated. The distance from the cemento- enamel junction to the alveolar crest of the posterior proximal surfaces was measured to the nearest 0.5 millimeter. Moreover, the appearance of the marginal bone was recorded.

For the separately recorded maxillary surfaces the mean distance was 0.9 or 1.0 mm. In the mandible, the distances amounted to 0.7 or 0.8 mm except for the mesial surface of the mandibular first premolar, where a distance of 0.5 mm was noted. At maxillary surfaces, 84-90% of the sites showed a distance of 1 mm or less and 94-100% a distance of 1.5 mm or less. At mandibular surfaces, 93-100% of the sites showed a distance of 1 mm or less and 100% a distance of 1.5 mm or less. At 4 sites located at the mesial surface of the first and second premolars in the maxilla, the distance was recorded as 2 mm. One site was classified as having an irregular image.

Based on these findings it was suggested that >2 mm would be the criterion of choice in epidemiological studies of bone loss in adole­scents.

Paper II

It is not known whether the periodontal health of adolescents has improved in recent decades, as has been the case with caries. In this study, the prevalence of marginal bone loss was compared in two groups of 16-year-old adolescents, born 1959 and 1972.

Bite-wing radiographs from 400 adolescents, 16 years of age in 1975 (born 1959), and from 400 adolescents, 16 years of age in 1988 (born 1972), i.e. approximately 40-45% of the respective annual cohort of adolescents in Umeå town, were randomly selected for analysis. The film used was the same (Kodak Ultraspeed) but between these years all X-ray equipment with a kV < 60 was replaced with equipment with a kV > 60. Whereas in 1975 the films were processed by hand, in 1988

21

all clinics used automatic developing machines. To be included in the analysis the radiographs were checked by an independent examiner for: available radiographs from the year the individual was 16 years old, acceptable technical quality, at least 50% of the proximal surfaces included in the analysis shown on the radiographs.

The radiographs were evaluated as described in Paper I. Based on the results in that study, marginal bone loss was recorded when the distance from the cemento-enamel junction to the alveolar crest exceeded 2 mm. In addition, presence of calculus, restorations and dental caries on proximal surfaces were recorded. A densitometric test was performed to decide whether a different quality of the radiographs would affect the radiographic contrast.

Approximately 10,000 surfaces were evaluated in each age cohort. No statistical differences in radiographic contrast were found between the two groups. Fourteen individuals (3.5%) in each age group displayed bone loss at one or more surfaces, with no difference between boys and girls. A majority of the sites with bone loss were found in the maxilla, most often at the mesial or distal surface of the first molar. Calculus and decayed or filled surfaces were significantly more common among the individuals born in 1959 compared to those born in 1972. No depend­ence was found between presence of decayed and filled proximal surfa­ces or calculus, respectively, and bone loss. The result of this study suggests that no changes in periodontal status, detectable on bite-wing radiographs, occurred in this population during the 13-year period.

Paper III

As the periodontal status in Swedish adolescents is hard to evaluate from existing studies, a cross-sectional study was performed in 1987 to describe the periodontal conditions in a group of adolescents from northern Sweden.

The total population of 16- and 18-year-olds in the county of Väster­

22

botten, Sweden, was divided into two strata, a rural and an urban. From the urban strata, represented by the town of Umeå, a random sample of approximately 10% of the total population was chosen. In the rural area, represented by the municipalities of Lycksele and Vilhelmina, a total examination was performed. 287 16-year-olds and 283 18-year-olds took part in a clinical examination including recording of plaque, bleed­ing on probing, calculus, probing attachment level and probing pocket depth; they also participated in an interview at which information was collected on occupation, use of tobacco, oral hygiene habits and the household's socioeconomic level. Four posterior bite-wing radiographs were taken and examined for the presence of bone loss and calculus.

As no differences in the clinical and radiographic parameters were found between the urban and the rural areas, the data were pooled for analysis. The overall prevalence of probing attachment loss at one or more sites in each individual (criterion ^ 2 mm) was 9.4% in 16-year- olds and 19.4% in 18-year-olds, with no significant differences between the sexes. The prevalence of attachment loss at one or more buccal/lingual surfaces was 6.6% in 16-year-olds and 15.5% in 18-year- olds and at proximal surfaces 2.8% and 4.9%, respectively. The 16- year-olds presented attachment loss at either buccal/lingual or proximal surfaces. Of the 18-year-olds, only 3 individuals had lost attachment at both proximal and buccal/lingual surfaces and of the rest either buccal/lingual or proximal surfaces were affected. Loss of attachment was more prevalent in the maxilla than in the mandible. In the 16-year- olds, the buccal and distal surfaces of the upper first molars were most frequently affected and in the 18-year-olds the buccal surface of the upper first molars and first premolars, followed by the distal surface of the upper first molars. In the individuals with attachment loss, the level of the attachment was situated in the majority of sites 2 or 3 mm from the cemento-enamel junction. The number of affected proximal sites per subject was 1-2 in all 16-year-olds, while of the 18-year-olds all but three had 1-2 sites affected. Bone loss was found in 1% of the subjects in both age groups (criterion > 2 mm).

Calculus was present in 32.7% of the 16-year-olds and in 44.9% of the

23

18-year-olds. Subgingival calculus was found in 3.1% and 9.2%, respectively. Pockets 2:4 mm were found in 5%. Toothbrushing was performed two or more times/day by 84.3% of the 16-year-olds and by 83.7% of the 18-year-olds, and once daily by the rest except for 1.4% and 2.5%, respectively, who brushed irregularly.

Paper IV

In the cross-sectional study (Paper III) two groups of individuals were distinguished, one with probing attachment loss at proximal surfaces and the other at buccal/lingual surfaces. The buccal/lingual loss was in fact located almost entirely buccally and is therefore in the following referred to as buccal attachment loss. Of 82 16- and 18-year-olds with attachment loss > 2 mm, only 3 had attachment loss on both proximal and buccal surfaces, possibly indicating that there might be different etiologies behind attachment loss at different surfaces. In addition to attachment loss caused by a bacterial inflammatory process, tissue trauma due to improper toothbrushing or, for instance, inadequately pla­ced orthodontic bands might account for an increasing proportion of cases with attachment loss among young people covered by an organized dental care system. In this study the groups with different localizations of compromised attachment were analyzed in relation to background variables.

The analyses included one group of individuals without probing attach­ment loss and one with probing attachment loss. The latter group consisted of one subgroup with proximal attachment loss and another with buccal attachment loss. The individuals from the two age groups were pooled and those with attachment loss at both proximal and buccal surfaces (n=3) were assigned to the group with attachment loss at proximal surfaces. The associations between attachment loss and the variables bleeding, plaque, supra- and subgingival calculus, probing pocket depth, oral hygiene habits, general health, medication, earlier orthodontic treatment with fixed appliance and socioeconomic level were tested, using the Chi-square test or Fischer's exact test.

24

No significant differences were found between the total group with attachment loss and the one without loss as regards general disease, medication, socioeconomic level, experience of orthodontic treatment, or presence of bleeding, plaque or supragingival calculus. However, in the subgroup with proximal loss, plaque, bleeding, calculus and pocket depths ^ 4 mm were more frequent than in the subgroup with buccal loss, while toothbrushing was more frequent in the latter subgroup than in the group without attachment loss. The differences between the sub­group with proximal attachment loss and that with buccal attachment loss indicate different etiologies of loss of tooth support at these ages.

Paper V

The results from Paper IV indicated that there might be different etiolo­gies for attachment loss at different surfaces. To further investigate possible background factors for the loss at proximal surfaces, a detailed description was obtained of the periodontal conditions of the individuals with proximal attachment loss. The group was followed for a 3-year period, during which all the individuals participated in an organized dental health care system.

All the 22 individuals identified earlier in the cross-sectional study (Paper III) as having attachment loss at proximal surfaces constituted the case group and 22 adolescents without attachment loss who also participated in the cross-sectional study were chosen at random as refe­rents, matched with the cases for age, sex and residential area. From the earlier study, information was obtained on probing pocket depth, pre­sence of plaque, bleeding on probing, calculus and radiographically assessed bone loss for each individual site with compromised attachment in the cases and for the corresponding sites in the referents. To gain further knowledge about the history of general disease and present and earlier use of medication, a new, more detailed interview was perform­ed. In addition, the dental records of all participants, in general cover­ing all dental visits from the age of 3 or 5, were collected. From these

25

records and from bite-wing radiographs taken regularly for caries diagnosis, information was obtained about orthodontic treatment, erup­tion disorders (ectopic eruption, infraocclusion etc.) and other clinical conditions that might influence the development of attachment loss.

One year after the cross-sectional study, the 22 cases and their referents were recalled for examination. This examination included recordings of the same clinical parameters as in the earlier examination and an inter­view on general health and medication. In addition, subgingival plaque samples were taken to determine presence of Actinobacillus actinomycetemcomitans (A. a.) and serum samples were taken to detect antibodies specific for leukotoxin from A.a. After another 3 years, the cases were examined once more, with a clinical examination and an interview, as well as sampling of subgingival plaque from sites with attachment loss and of serum. In connection with these examinations, subgingival calculus was removed in order to perform measurements of probing attachment level. After the 1-year examination, the dentists of all individuals who exhibited attachment loss were personally informed about the periodontal state of their patients. Treatment, if needed, was given and decisions regarding institution of individualized prophylactic programs or specialist consultation were made by the patient's own dentist.

According to the dental records, cases and referents had had access to dental care to a similar extent since preschool age. Four of the cases had been observed by their dentists to have poor oral hygiene and/or periodontal pockets and therefore to need prophylactic treatment. These included the two most severe cases, with 4 and 5 sites affected at the cross-sectional examination. Atopic diseases and frequent general infec­tions were evenly distributed between cases and referents. One boy in the case group had Mb Hodgkin. Bleeding on probing at sites with attachment loss was found in 12 of the cases and in 4 of the referents at the corresponding sites. Plaque was almost equally common at the ac­tual sites in the two groups, as was the presence of supragingival calcu­lus. Subgingival calculus was found in 6 of the cases and in none of the referents. Orthodontic bands were almost equally common at the actual

26

sites in the two groups. At one of the sites with attachment loss a defec­tive filling had been present and another had a fissure in the root sur­face. Also, at one site with attachment loss an inftaocclusion of a neigh­boring primary tooth had been present earlier and at one site the second premolar had erupted in close contact with the mesial surface of the first molar. In the referents no defective fillings or eruption disorders were found at any of the corresponding sites.

One individual in the case group had moved abroad, leaving 21 cases and 21 referents for the re-examinations. At the re-examinations 10 of the 21 earlier identified cases did not meet the criterion of one or more sites with a probing attachment level of ^ 2 mm. None of the cases showed progression ^ 2 mm of the lesions. Oral hygiene, measured as number of sites with plaque, improved in the cases over the three years. Four of the cases harbored A.a. at affected sites in year 1 and three of these in year 3. Eight of the cases and four referents had antibodies specific for Æa.-leukotoxin. The two most severe cases, in terms of number of sites with attachment loss, showed bleeding at one or more of the sites, harbored A.a. at these sites and had serum titers against A.a.- leukotoxin. In one of these the number of sites increased during the period.

Various factors may obviously contribute to proximal attachment loss. The majority of the 22 cases had solitary affected sites, some without signs of inflammation, possibly indicating that the destruction of the periodontium had occurred earlier and not neccesarily as a result of inflammatory periodontal disease. Two cases with several affected sites were found and these had poor oral hygiene and gingival inflammation at the affected sites. They also harbored A.a. and had serum-titers of antibodies against leukotoxin from A.a. The clinical findings may imply an active periodontitis in these individuals, thus giving a prevalence of periodontitis of 0.3% in the population.

27

Paper VI

A variety of etiological factors for gingival recession have been suggest­ed in the literature (for review see Woofter 1969, Paterson 1979, Watson 1984). The relationship between single etiologic factors and recession has been studied but the relative importance of these factors has not been investigated. This study focused on the association between buccal attachment loss and the factors earlier suggested as etiological for gingival recession.

A group of 18-year-olds, identified in the cross-sectional study two years earlier (Paper III) as having buccal attachment loss à 1 mm, and randomly selected referents identified without buccal attachment loss, were examined, interviewed and observed in this study. In the clinical examinations the following parameters were recorded: high frenum attachment, presence of plaque, bleeding on probing, calculus, probing attachment level and width of the attached gingiva.

In the interview, the following information was gathered: brand of toothbrush and toothpaste, bristle hardness, frequency of brushings, if instruction in toothbrushing technique had been received, use of smoke­less tobacco and habits which may result in trauma to the gingiva.

The subjects were asked to brush their teeth in their normal way and this was observed while the following parameters were registered: hand used for brushing, amount of toothpaste, where in mouth the brushing began, type of toothbrushing technique, light or strong brushing force, time spent on brushing, toothbrushing performance.

At the clinical examination impressions were taken of the teeth and gums and plaster casts were made. These casts were used to assess the type of occlusion, buccal alignment of molars, premolars and canines as well as the width of the buccal alveolar tissue. The dental records for each subject were also checked for previous orthodontic diagnoses and orthodontic treatment.

28

The case group consisted of 71 and the referent group of 66 subjects. The case group comprised two subgroups, one identified as having buccal attachment loss in 1987 and the other with attachment loss occurring in the years 1987-89. In total, information on 28 variables was collected and tested for association with buccal attachment loss, using Chi-square test, logistic regression analysis and variance compo­nents analysis.

The average number of affected sites was 4.4 in the case group, 5.6 in the subgroup identified with buccal attachment loss in 1987 and 2.7 in the subgroup where the loss had occurred in 1987-89. Of the adolescents identified with buccal attachment loss in 1987, 31 (76%) showed a higher number of affected sites in 1989 and 16 (36%) showed progression ( ^ 1 mm) of one or more sites during the period.

The factors identified in the statistical analyses as being related to bucc­al attachment loss were thin buccal alveolar tissue, narrow width of the attached gingiva and presence of teeth with buccal displacement. The re­sults indicate that the anatomy of the buccal alveolar process is related to buccal attachment loss in populations where the level of oral hygiene is high.

DISCUSSION

The cross-sectional study (Paper III) showed an overall prevalence of attachment loss (criterion ^ 2 mm) of about 9% in 16-year-olds and 19% in 18-year-olds. The majority of the sites with attachment loss were found at buccal surfaces, i.e. in 6.6% of the 16-year-olds and in 15.5% in the 18-year-olds. The prevalence of proximal attachment loss was 2.8% and 4.9%, respectively. These findings, based on a full mouth examination of four sites per tooth, can be compared with two recent studies of periodontal conditions using probing attachment level as the measure of compromised attachment. In a report from the

29

"National Survey of Oral Health in Schoolchildren" performed in the US 1986-87, Bhat (1991) described results on periodontal health based on examinations of 11,111 children aged 14 to 17 years. All mesial and buccal surfaces were examined in that study for bleeding, calculus and attachment loss. The examining methods were the same as in our study. The results showed that 21.7% had attachment loss (criterion ^ 2 mm) at any site and that the prevalence of recession (criterion >2 mm) was 1.5%. In Amsterdam, Netherlands, 4,565 15-16-year-olds were studied by examining proximal surfaces of first molars and incisors (van der Velden et al. 1989). The prevalence of probing attachment loss with the criterion ^ 1 mm was 5%; recalculated for the criterion ^ 2 mm the prevalence was 4.2%. Although the partial examination might underesti­mate the prevalence of attachment loss compared to the full mouth examination (Kingman et al. 1988), it seems that the prevalence of proximal attachment loss in Amsterdam adolescents resembles that of the present sample of Swedish adolescents. The US national survey, where both buccal and proximal surfaces were examined, shows a some­what higher figure for the overall prevalence compared to the Swedish sample, while the prevalence of buccal attachment loss is considerably lower in the US.

The surface most prone to show probing attachment loss was the buccal of the upper first molar in the Swedish adolescents. For proximal sur­faces the distal surface of the first maxillary molar was the most affected (Paper III). This agrees with the US National survey of schoolchildren (Bhat 1991), where the maxillary molars were reported to be the most common teeth for attachment loss. The intraoral distribution of proximal attachment loss in a group of Navajo Indians 14-19 years of age also coincides with the present Swedish findings (Wolfe & Carlos 1987).

In the Swedish sample subgingival calculus at one or more sites per subject was found in 3% and 9%, respectively, in the two age groups compared to 23% in the US survey (Bhat 1991). The subgingival calcu­lus was in our sample more often found in subjects with proximal attachment loss than in those without loss or with buccal attachment

30

loss. Of the group with proximal loss, 36% had subgingival calculus compared with 5% of those with buccal attachment loss (Paper IV). In a longitudinal study of proximal attachment loss in English adolescents, the presence of subgingival calculus at baseline correlated significantly with the 5-year increment of attachment loss (Clerehugh et al. 1990). In a recent longitudinal study by Ånerud et al. (1991) the natural history and clinical course of calculus formation is described in two cohorts of men from Sri Lanka and Norway, respectively. The influence of calcu­lus on attachment loss was very slight in the Norwegian sample, while in the Sri Lanka sample individuals with subgingival calculus lost more periodontal attachment.

A most striking finding in the Swedish sample was that the buccal sur­faces had much more attachment loss than the proximal surfaces (Paper III). Moreover, two distinct groups of individuals, with either buccal or proximal attachment loss, could be distinguished, a finding that might indicate a different etiology for the attachment loss. In view of findings that young individuals with only buccal attachment loss had fewer oral deposits and a lower level of gingivitis than individuals with buccal and mesiobuccal attachment loss, Clerehugh et al. (1988) suggested two different etiologies for loss of attachment, the loss in the first group being ascribed to trauma from toothbrushing and in the second group to early periodontitis. This suggestion by Clerehugh and co-workers was based on a longitudinal study of the natural history of early periodontal disease in adolescents selected to have a high risk of developing loss of attachment (1986, 1988, 1990). In the present study of Swedish adole­scents another approach is used. The result of the cross-sectional study was used to form groups with different locations of attachment loss and these adolescents were investigated separately by means of case-referent studies to gain more knowledge of the association between presence of attachment loss and various background factors. This analytical ap­proach presented in paper IV resulted in the identification of statistically significant differences between the groups with proximal attachment loss and buccal attachment loss, respectively. Plaque, bleeding, calculus and pocket depths ^ 4 mm were found more often in the group with proximal attachment loss compared to the group with buccal attachment

31

loss, while toothbrushing was more frequent in the latter group than in the group without attachment loss. Thus, the results were in accordance with the findings of Clerehugh et al. (1988).

Several reasons for attachment loss could, however, be found for both proximal and buccal attachment loss and many of these factors might work at a single site or individual. As these factors may have been hid­den in the group means, an analysis was performed of each subject in the group with proximal attachment loss (Paper V). The results showed that not only were some losses not connected to present inflammation but some could also possibly be ascribed to earlier eruption and developmental disorders as well as to local irritants such as overhanging fillings.

Buccal attachment loss was quite common in this population. When the criterion of probing attachment level ^ 1 mm from the cemento-enamel junction was applied, 22.5% of the 16-year-olds were identified in the cross-sectional study. Our results (Paper IV), supported by those from the study by Clerehugh et al. (1988), show that this loss of attachment in adolescents is most likely not associated with the background factors usually connected to inflammatory periodontal disease. Instead, the results of the case-referent study (Paper VI) showed that in this popula­tion of adolescents, who exhibit good oral hygiene, the attachment loss is associated with the anatomy of the buccal alveolar process. Buccal alignment of teeth or a thin buccal alveolar tissue would possibly pre­dispose for the development of buccal attachment loss. These findings agree essentially with the findings by Löst (1984) of a correlation between presence of gingival recession and bone dehiscence depth as­sessed after surgical exploration. To evaluate these possible risk factors, prospective epidemiological or experimental studies are needed.

An interpretation of our studies (Papers III-VI) would be that the loss of attachment found in the cross-sectional study (9% and 19% respectively) occurs in the majority of the cases at buccal surfaces and has a different etiology from the loss found at proximal surfaces, seemingly not connected to the factors usually associated with periodon-

32

tal disease. Also, the proximal attachment loss could be ascribed to a varying etiology. A tentative conclusion was that only 0.3% of the adolescent population in northern Sweden had periodontitis.

This figure, 0.3%, is close to levels of juvenile periodontitis found in populations comparable to this Swedish sample. However, the signs of periodontal disease in the two most severe cases identified in Paper V did not fit with the definition of Baer (1971). Still, definitions of juve­nile periodontitis do differ in studies of the disease. The results from earlier studies on prevalence of compromised attachment (Table 2) dif­fer from the reports on the prevalence of juvenile periodontitis. Our findings of a varying etiology of the loss of tooth support indicate that the groups of individuals with compromised attachment listed in Table 2 include individuals with active periodontitis as well as individuals with earlier experienced disease or with defects caused by non-inflammatory trauma.

The statistical methods used for testing hypotheses or the strength of associations in studies of periodontal disease has been thoroughly discussed (Sterne et al. 1990, Hujoel et al. 1990a,b, Emrich 1990). The main problem is that variables used in periodontal research are often both site- and subject-based, thus raising the issue of whether or not the site-based clinical measurements should be handled as subject-based statistics and whether different approaches may give differing inferences. In the analysis of the cross-sectional study (Paper IV), a subject-based approach was used as the aim was to detect differences in background factors at group level. The statistical analysis was restricted to a Chi-square test. In the case-referent study of buccal attachment loss (Paper VI) the analyses were performed in three steps. The Chi-square test was primarily used in order to make comparisons with other stud­ies, the logistic regression provided assessments of the effect of combinations of subject-based exposure variables and the variance components analysis examined the relations between site-based exposure variables and buccal attachment loss without requiring assumptions of independence between sites and subject. Due to the low number of individuals with proximal attachment loss and the low number of

33

proximal sites per individual with loss, no statistical analysis was per­formed in Paper V. This problem of a low number of affected sites/individuals actually applied to many of the analyses in these studies because the large number of sites without loss cause major dilution of the effects and obscure the analysis or make inferences difficult. This problem, obviously inherent in studies of periodontal conditions, especi­ally in populations with a low prevalence and severity of these condi­tions, will constitute a problem also in further investigations of perio­dontal conditions in adolescents.

As pointed out in the introduction, probing measurements of compromi­sed attachment have been discussed and tested for validity and reproducibility. According to Badersten et al. (1984), approximately 90% of the recordings of the probing attachment level could be repro­duced to within +1.5 mm. Moreover, they showed that the reproducibility varied significantly between tooth types, surfaces, and probing pocket depths; buccal surfaces and shallow pockets having less variability. These results stem from a probing technique where probably a direct measure of attachment loss was used, in contrast to the indirect technique in the present studies. The indirect technique, in which the attachment level is calculated with values rounded down, might have even less reproducibility; a conservative criterion for attachment loss, probing attachment level > 2 mm, was therefore chosen in these studies. For the measurements at buccal surfaces the criterion chosen was ^ 1 mm as the reproducibility is greater at these easily accessable, shallow pockets. The problem of the validity of measurements of probing attach­ment level is evident in the results described in Paper V. Only half of the sites with proximal attachment loss identified in the cross-sectional study were re-identified at the reexaminations after 1 and 3 years. This was probably because calculus had been removed at the examinations so that, by reducing the degree of gingival inflammation, the tissue tonus was increased, changing the ability of the probe to penetrate the crevice (Listgarten 1980).

Radiographic assessment of bone loss was used in two groups of 16- year-olds (Paper II) and in the cross-sectional study (Paper III). In order

34

to establish criteria for bone loss in adolescents, a study was performed on the normal appearance of the alveolar bone in the posterior areas (Paper I). The normal range of the distance between the cemento-enamel junction and the alveolar crest was within 0-2 mm and for epidemiologi­cal studies a threshold of > 2 mm was suggested. A test of intra-exami­ner reproducibility of measurement of the bone level revealed that 99% of the readings were reproduced within ±0.5 mm. While this seems satisfactory, we experienced a need for several sessions of calibration in order to achieve good inter-examiner reproducibility (97.5%, Paper II). This illustrates the difficulties of comparing results of bone loss in different studies even if the criteria and methods are alike. However, the reproducibility test was based on reexaminations of radiographs, and thus did not include the elements of variations in radiographic tech­nique.

The comparison of the 1975 and 1988 cohorts of 16-year-olds showed the same prevalence (3.5%) of bone loss in the two groups. Although this method is less sensitive compared to clinically assessed attachment loss, the results from the comparison of the two groups of adolescents, 13 years apart, indicate that the prevention practiced in the dental health system during this time did not change the prevalence of compromised attachment at proximal surfaces. The preventive program is based on dietary counselling, instruction in toothbrushing and use of fluorides. It could be that other preventive measures may be needed to improve periodontal health. It is also conceivable that the preventive program available for the 16-year-olds in 1975 was so effective that a further improvement could be achieved only with very rigorous prevention. Moreover, preventive measures cannot influence other factors which may contribute to periodontal breakdown, e.g., defective fillings, root fissures and eruption disorders.

Public health programs initiated to control periodontal diseases have been based on the assumption that gingivitis progresses to periodontitis, so that gingivitis is considered worth preventing. Programs designed to improve oral hygiene, based on education and professional toothclean- ing, have been shown to have a great impact in this respect and on the

35

prevalence of gingivitis (Axelsson & Lindhe 1974, 1975, Lindhe et al. 1975, Axelsson et al. 1976, Kjaerheim et al. 1980). Moreover, oral hy­giene levels are improving over time (Douglass et al. 1983) and for adolescents in northern Sweden it has been shown that these levels as well as the use of oral hygiene measures have improved (Källestål et al. 1990).

The so called specific plaque theory, presented by Loesche in 1976, suggested that one or only a few types of microorganisms are responsi­ble for the development of periodontitis. According to this theory, preventive programs should include a microbiological diagnosis and use the appropriate antimicrobial compounds to eliminate the pathogens. Moreover, the search for predictors of periodontal diseases is motivated by the possibility of detecting a risk group on which prevention and treatment could be concentrated. These approaches envisage screening for periodontal diseases. However, there are certain prerequisites for a screening procedure. The disease should be regarded as an important health problem, effective and acceptable treatments should be available, methods for screening should be available, early treatment should affect the prognosis, the natural history of the disease should be adequately understood and there should be an agreed policy on whom to treat (Wilson & Jungner 1968, Sheiham 1978). These conditions are hardly fulfilled by periodontal diseases in adolescents. Although the perception of periodontal diseases has changed a great deal, there are still no pre­cise predictors available and the hypothesis of a specific agent for the disease has not won general agreement. Thus, it is still the concept of regular removal of plaque, the target being gingivitis rather than periodontitis, that dominates the public health measures for preventing periodontitis.

To implement preventive measures for a disease with a prevalence of < 1 % in adolescents might not seem justified. Moreover, the lack of clear risk factors rules out the identification of risk groups and shifts the question of prevention from primary prevention to secondary, i.e. treat­ment when early signs of disease are present. Not even this path of action is easy in the absence of a diagnostic measure of active disease.

36

All in all, the scope for prevention is not very large. On the other hand, the Swedish dental health system already offers a great deal of preven­tion to children and adolescents; the prevention of periodontitis is not a matter of implementing new preventive measures but rather of directing existing efforts at those in most need. Also, while the prevalence of periodontitis is low, the cost, both economically and socially, for treat­ing severe periodontal disease in adults is high and thus, if prevention could be successful in young individuals it might be justified, in that individuals who already have attachment loss run a greater risk of experiencing progressive loss (Haffajee et al. 1991). These questions of priorities in dental health care are, however, more relevant for political and economic considerations than for health science. The results of this thesis on periodontal conditions in adolescents make it possible to discuss strategies for prevention of attachment loss in young individu­als, if such prevention is given priority.

The primary issue in secondary prevention is how to identify individuals with early signs of compromised attachment. The studies in this thesis clearly indicate that a difference in etiology is likely for attachment loss at buccal and proximal surfaces, respectively. Thus, for identification of proximal attachment loss, if a low sensitivity is accepted and posterior bite-wing radiographs are available, assessment of presence of bone loss could serve as a screening procedure. The shortcomings of this method are that early signs of disease are neglected and only the posterior teeth are covered. Another approach might be partial examination of pocket depths, suggested in Paper V. As shown in that paper, pocket depth ^ 4 mm was common at the proximal sites with attachment loss. If a partial recording of pocket depths, covering the mesial and distal surfaces of the first and second molars and incisors, had been performed, 86% of the cases with proximal loss would have been identified in the present sample of adolescents.

These identified cases with attachment loss would then be examined and could all be followed to assess disease activity by measuring progress of attachment loss. Treatment would be given only to those with active disease. An alternative, also discussed in Paper V, may be to treat all

37

those identified with compromised attachment if they also exhibit bleed­ing and/or calculus at the site with attachment loss.

If this strategy of screening by means of a partial recording of pocket depth is used, an assessment of the status at the buccal surface of the first molar could be added to identify early signs of buccal attachment loss.

Although the etiology of periodontal disease is still obscure and the available diagnostic methods do not have the ability to identify active disease, the present knowledge of the distribution, inter and intra- individually, and of clinical signs connected with periodontal disease, makes it possible to develop a preventive strategy that might be more effective than the present preventive measures directed at gingivitis.

38

CONCLUSIONS

The main findings of the studies were:► In adolescents without any sign of gingival inflammation or loss of

probing attachment the distance between the cemento-enamel junc­tion and the alveolar crest was in the 0-2 mm range; thus, for epidemiologic purposes a distance of > 2 mm between the alveolar crest and the cemento-enamel junction seems appropriate as a crite­rion of bone loss in adolescents.

► The same prevalence of bone loss in 16-year-olds (3.5%) was found in 1975 and 1988.

► In the cross-sectional study the prevalence of probing attachment loss was 9% in 16-year-olds and 19% in 18-year-olds. The compromised attachment was found mainly at buccal surfaces. Proximal loss of attachment was found in 3% and 5% of the individuals, respectively.

► The group with proximal attachment loss had more plaque, bleeding, calculus and pocket depths ^ 4 mm than the group with buccal loss. In the latter group, toothbrushing was more frequent than in the group without attachment loss. The results indicate different etiolo­gies of loss of tooth support at different tooth-surfaces at these ages.

► Various factors may contribute to the loss of proximal attachment in young individuals. A majority of the individuals with attachment loss had solitary affected sites, some without signs of inflammation, possibly indicating that the destruction of the periodontium had occurred earlier. Two cases were tentatively given the diagnosis periodontitis, thus giving a prevalence of 0.3% in this population.

► Factors related to the anatomy of the alveolar process seem to be associated with buccal attachment loss.

39

ACKNOWLEDGEMENTS

I wish to express my sincere gratitude to all who have assisted me and helped me to accomplish this thesis. In particular, I am indebted to:

Dr Lars Matsson, not only for his knowledge in Periodontology that he has shared with me, but also for teaching me the value of a meticulous research method and for always showing interst in and allocating time to my work.

Professor Anna-Karin Holm for always generously accepting me in her department; without her openness and dynamic mind I would never have stayed long enough to finish this thesis.

Assistants Annmarie Ljus-Andersson and Ingegerd Andersson for their excellent, accurate work and for good companionship and loyalty during our tiresome trips around the county.

Ulla Holmgren, for patience with the new word-processing program, for good advice concerning the layout of this thesis but most of all for being a never-failing friend.

All staff at the Public Dental Health Clinics for their readiness and helpfulness during my visits.

My co-authors DDS Sten Persson and BSc Stig Uhlin for generous help, hard and fruitful discussions and many a good laugh.

Friends and colleagues at the Dental School, Umeå, for help with pho­tos, radiographic and immunologic methods in practice and theory, stimulating discussions, advice etc, but most of all a warm and friendly attitude during all the grey days.

Professor Brian Burt and his staff and students at the Dental Health Program, School of Public Health, Ann Arbor, Michigan, for receiving

40

me in the department, allowing me to stand back from my own work and fruitful discussions.

Mr Patrick Hort for prompt revision of the English.

The Swedish Institute for the scholarship enabling me to go to Michigan and to the Faculty of Odontology for permanent support during these studies.

The Journal of Clinical Periodontology for allowing me to copy articles not yet published.

This thesis was supported by grants from the Faculty of Odontology, University of Umeå, the Swedish Dental Society and the Joint Com­mittee of the North Health Region, Sweden.

Last but not least, all adolescents taking part in my studies.

41

REFERENCES

Aass AM, Albandar J, Aasenden R, Tollefsen T, Gjermo P. Variation in prevalence of radiographie alveolar bone loss in subgroups of 14-year-old schoolchildren in Oslo. J Clin Periodontol 1988;15:130-3.Ainamo J, Barmes D, Beagrie G, Cutress T, Martin J, Sardo-Infirri J. Development of the World Health Organization (WHO) community periodontal index of treatment needs (CPITN). Int Dent J 1982;32:281-91.Ainamo J, Paloheimo L, Nordblad A, Murtomaa H, Gingival recession in schoolchil­dren at 7, 12 and 17 years of age in Espoo, Finland. Community Dent Oral Epidemiol 1986;14:283-6.Ånerud KE, Robertson PB, Löe H, Ånerud Å, Boysen H, Patters MR. Periodontal dis­ease in three young adult populations. J Periodontal Res 1983;18:655-68.Ânerud Â, Löe H, Boysen H. The natural history and clinical course of calculus forma­tion in man. J Clin Periodontol 1991;18:160-70.Asikainen S. Occurrence of Actinobacillus actinomycetemcomitans and spirochetes in relation to age in localized juvenile periodontitis. J Periodontol 1986;57:537-41. Asikainen S, Jousimies-Somer H, Kanervo A, Summanen P. Certain bacterial species and morhpotypes in localized juvenile periodontitis and in matched controls. J Perio­dontol 1987;58:224-30.Axelsson P, Lindhe J . The effect of a preventive programme on dental plaque, gingivi­tis and caries in schoolchildren. Results after one and two years. J Clin Periodontol 1974;1:126-38.Axelsson P, Lindhe J. Effect of fluoride on gingivitis and dental caries in a preventive program based on plaque control. Community Dent Oral Epidemiol 1975;3:156-60. Axelsson P, Lindhe J, Wäseby J. The effect of various plaque control measures on gingivitis and caries in schoolchildren. Community Dent Oral Epidemiol 1976;4:232-9. Bäckman N, Holm A-K, Keskitalo E, Gustafsson L. Dental health of national ser­vicemen in Umeå. Swed Dent J 1981;5:129-39.Bader sten A, Nilvéus R, Egelberg J. Reproducibility of probing attachment level measurements. J Clin Periodontol 1984;11:475-85.Bader sten A, Nilvéus R, Egelberg J. Effect of nonsurgical periodontal therapy. V. Pat­terns of probing attachment loss in non-responding sites. J Clin Periodontol 1985;12:270-82.Baelum V, Fejerskov O, Karring T. Oral hygiene, gingivitis and periodontal breakdown in adult Tanzanians. J Periodontal Res 1986;21:221-32.Baelum V, Fejerskov O, Manji F. Periodontal diseases in adult Kenyans. J Clin Periodontol 1988;15:445-52.Baer PN. The case for periodontosis as a clinical entity. J Periodontol 1971;42:516-9. Bay I. Mundhygiejne og parodontale forhold hos en gruppe Sjaellandske vaemeplig- tige. Tandlaegebladet 1972;76:1176-86.Benjamin SD, Baer PN. Familial patterns of advanced alveolar bone loss in adolescence (periodontosis). Periodontics 1967;5:82-8.Bhat M. Periodontal health of 14-17-year-old US schoolchildren. J Pubi Health Dent 1991;51:5-11.

43

Bille J, Hesselgren K, Thylstrup A. Dental caries in Danish 7-, 11- and 13-year-old children in 1963, 1972 and 1981. Caries Res 1986;20:534-542.Birkeland JM, Bragelien J . Continual highly significant decrease in caries prevalence among 14-year-old Norwegians. Acta Odontol Scand 1987;45:135-40.Björn A-L, Andersson U, Olsson A. Gingival recession in 15-year old pupils. Swed Dent J 1981;5:141-6.Björn A-L, Hailing A. Clinical and roentgenographic examination of 697 Swedish re­cruits, year group 1980. Swed Dent J 1983;7:129-39.Blankenstein R, Murray JJ, Lind OP. Prevalence of chronic periodontitis in 13-15-year- old children. A radiographic study. J Clin Periodontol 1978;5:285-92.Blomlöf L, Hammarström L, Lindskog S. Occurrence and appearance of cementum hypoplasias in localized and generalized juvenile periodontitis. Acta Odontol Scand 1986;44:313-20.Boyle Jr WD, Via Jr WF, McFall Jr WT. Radiographic analysis of alveolar crest height and age. J Periodontol 1973;44:236-43.Budtz-Jörgensen E, Ellegaard J, Ellegaard B, Jörgensen F, Kelstrup J. Cell-mediated immunity in juvenile peridontitis and levamisole treatment. Scand J Dent Res 1978;86:124-9.Burt BA, Ismail Al, Eklund SA. Periodontal disease, tooth loss, and oral hygiene among older Americans. Community Dent Oral Epidemiol 1985;13:93-6.Capilouto ML, Douglass CW. Trends in the prevalence and severity of periodontal diseases in the US: A public health problem? J Pubi Health Dent 1988;48:245-51. Carlos JP, Wolfe MD, Kingman A. The extent and severity index: a simple method for use in epidemiologic studies of periodontal disease. J Clin Periodontol 1986;13:500-5. Caton J. Periodontal diagnosis and diagnostic aids. Proceedings of the World Work­shop in Clinical Periodontics. Princetown, New Jersey; 1989 July 23-27.Christer son LA, Slots J, Zambon JJ, Genco RJ. Transmission and colonization of Actinobacillus actinomycetemcomitans in localized juvenile periodontitis patients. J Periodontol 1985;56:127-31.Clerehugh V, Lennon MA. A two-year longitudinal study of early periodontitis in 14- to 16-year-old schoolchildren. Community Dent Health 1986;3:135-41.Clerehugh V, Lennon MA, Worthington HV. Aspects of the validity of buccal loss of attachment ^ 1 mm in studies of early periodontitis. J Clin Periodontol 1988;15:207- 10.Clerehugh V, Lennon MA, Worthington HV. 5-year results of a longitudinal study of early periodontitis in 14- to 19-year-old adolescents. J Clin Periodontol 1990;17:702-8. Crossner C-G, Unell L. A longitudinal study of dental health and treatment need in Swedish teenagers. Community Dent Oral Epidemiol 1986;14:10-4.Curtis MA, Gillett IR, Griffiths GS, Maiden MFJ, Sterne JAC, Wilson DT et al. Detec­tion of high-risk groups and individuals for periodontal diseases: laboratory markers from analysis of gingival crevicular fluid. J Clin Periodontol 1989;16:1-11.Davies PHJ, Downer MC, Lennon MA. Periodontal bone loss in English secondary school children. A longitudinal radiological study. J Clin Periodontol 1978;5:278-84. Douglass CW, Gillings D, Sollecito W, Gammon M. National trends in the prevalence and severity of the periodontal diseases. J Am Dent Assoc 1983;107:403-12.

44

Emrich LJ. Common problems with statistical aspects of periodontal research papers. J Periodontol 1990;61:206-8.Fischman SL. Current status of indices of plaque. J Clin Periodontol 1986;13:371-4. Flores-de-Jacobry L, Schoop S, Weichster C, Zafiropoulos G-G. Periodontal conditions in Hesse, Federal Republic of Germany, measured by CPITN. Community Dent Oral Epidemiol 1989;17:307-9.FourelJ. Periodontosis. A periodontal syndrome. J Periodontol 1972;43:240-55. Frentzen M, Pßffle W, Nolden R. Gingivarezessionen bei jungen Erwachsenen als Folge einer intensiven Zahnpflege? Dtsch Zahnärztl Z 1989;44:373-4.Gjermo P, Bellini HT, Santos VP, Martins JG, Ferracyoli JR. Prevalence of bone loss in a group of Brazilian teenagers assessed on bite-wing radiographs. J Clin Periodontol 1984;11:104-13.Goodson JM. Clinical measurements of periodontitis. J Clin Periodontol 1986; 13:446- 55.Goodson JM, Tanner ACR, Haffajee AD, Somber g er GC, Socransky SS. Patterns of progression and regression of advanced destructive periodontal disease. J Clin Perio­dontol 1982;9:472-81.Gorman WJ. Prevalence and etiology of gingival recession. J Periodontol 1967;38:316- 22.

Griffiths GS, Wilton JMA, Curtis MA, Maiden MFJ, Gillett IR, Wilson DT et al. Detec­tion of high-risk groups and individuals for periodontal diseases. Clinical assessment of the periodontium. J Clin Periodontol 1988;15:403-10.Haffajee AD, Socransky SS, Goodson JM. Comparison of different data analyses for detecting changes in attachment level. J Clin Periodontol 1983;10:298-310.Haffajee AD, Socransky SS, Lindhe J, Kent RL, Okamoto H, Yoneyama T. Clinical risk indicators for periodontal attachment loss. J Clin Periodontol 1991;18:117-25.Hailing A, Björn A-L. Periodontal status in relation to age of dentate middle aged wo­men. A 12 year longitudinal and a cross-sectional population study. Swed Dent J 1986;10:233-42.Hansen BF, Bjertness E, Gjermo P. Changes in periodontal disease indicators in 35- year-old Oslo citizens from 1973 to 1984. J Clin Periodontol 1990;17:249-54.Hansen BF, Gjermo P, Bergwitz-Larsen KR. Periodontal bone loss in 15-year-old Norwegians. J Clin Periodontol 1984;11:125-31.Hoover JN, Ellegaard B, Attström R. Periodontal status of 14-16-year-old Danish schoolchildren. Scand J Dent Res 1981a;89:175-9.Hoover JN, Ellegaard B, Attström R. Radiographic and clinical examination of perio­dontal status of first molars in 15-16-year-old Danish schoolchildren. Scand J Dent Res 1981b;89:260-3.Hoover JN, Tynan JJ. Periodontal status of a group of Canadian adults. J Canad Dent Ass 1986;52:761-3.Hugoson A, Jordan T. Frequency distribution of individuals aged 20-70 years accord­ing to severity of periodontal disease. Community Dent Oral Epidemiol 1982; 10:187- 92.Hugoson A, Koch G, Rylander H. Prevalence and distribution of gingivitis-periodontitis in children and adolescents. Epidemiological data as a base for risk group selection. Swed Dent J 1981;5:91-103.

45

Hujoel PP, Loesche WJ, DeRouen TA. Assessment of relationships between site-speci­fic variables. J Periodontol 1990a;61:368-72.Hujoel PP, Moulton LH, Loesche WJ. Estimation of sensitivity and specificity of site- specific diagnostic tests. J Periodont Res 1990b;25:193-6.Hull PS, Hillam DG, Beal JF. A radiographic study of the prevalence of chronic periodontitis in 14-year-old English schoolchildren. J Clin Periodontol 1975;2:203-10. lmrey PB. Considerations in the statistical analysis of clinical trials in periodontitis. J Clin Periodontol 1986;13:517-28.Ismail Al, Morrison EC, Burt BA, Caffesse RG, Kavanagh MT. Natural history of periodontal disease in adults. Findings from the Tecumseh periodontal disease study, 1959-87. J Dent Res 1990;69:430-5.Johnson NW, Griffiths GS, Wilton JMA, Maiden MFJ, Curtis MA, Gillett IR et al. Detection of high-risk groups and individuals for periodontal diseases. Evidence for the existence of high-risk groups and individuals and approaches to their detection. J Clin Periodontol 1988;15:276-82.Jorgensen RJ, Levin LS, Hutchersson ST, Salinas CF. Periodontosis in sibs. Oral Surg, Oral Med, Oral Pathol 1975;39:396-402.Jorkjend L, Birkeland JM. Alveolar bone loss in the permanent first molars of Norwe­gian schoolchildren receiving systematic dental care. Community Dent Oral Epidemiol 1976;4:22-4.Källestål C, Holm A-K, Ollinen P. Dental health in 13-year-olds in Västerbotten county, Sweden. Changes over twenty years. Swed Dent J 1990;14:193-200.Kingman A, Morrison E, Löe H, Smith J. Systematic errors in estimating prevalence and severity of periodontal disease. J Periodontol 1988;59:707-13.Kjaerheim V, von der Fehr FR, Pouls en S. Two-year study on the effect of professional toothcleaning on schoolchildren in Oppegård, Norway. Community Dent Oral Epide­miol 1980;8:401-6.Koch G. Evidence for declining caries prevalence in Sweden. J Dent Res 1982;61:Spec issue: 1340-5.Kornman KS. Nature of periodontal diseases. Assessment and diagnosis. J Periodontal Res 1987;22:192-204.Kronauer E, Borsa G, Lang NP. Prevalence of incipient juvenile periodontitis at age 16 years in Switzerland. J Clin Periodontol 1986;13:103-8.Latcham NL, Powell RN, Jago JD, Seymour GJ, Aitken JF. A radiographic study of chronic periodontitis in 15 year old Queensland children. J Clin Periodontol 1983;10:37-45.Lehner T, Wilton JMA, Ivanyi L, Manson JD. Immunological aspects of juvenile periodontitis (periodontosis). J Periodont Res 1974;9:261-72.Lennon MA, Davies RM. Prevalence and distribution of alveolar bone loss in a popula­tion of 15-year-old schoolchildren. J Clin Periodontol 1974; 1; 175-82.Lervik T, Riordan PJ, Haugejorden O. Periodontal disease and approximal overhangs on amalgam restorations in Norwegian 21-year-olds. Community Dent Oral Epidemiol 1984;12:264-8.

46

Lindhe J, Axelsson P, Tollskog G. Effect of proper oral hygiene on gingivitis and dental caries in Swedish schoolchildren. Community Dent Oral Epidemiol 1975;3:150- 5.Lindhe J, Haffajee AD, Socransky SS. Progression of periodontal disease in adult sub­jects in the absence of periodontal therapy. J Clin Periodontol 1983;10:433-42.Lindskog S, BlomlcfL. Cementum hypoplasia in teeth affected by juvenile periodontitis.J Clin Periodontol 1983;10:443-51.Listgarten MA. Periodontal probing. What does it mean? J Clin Periodontol 1980;7:165-76.Listgarten MA. Pathogenesis of periodontitis. J Clin Periodontol 1986;13:418-25. Listgarten MA. Why do epidemiologic data have no diagnostic value? In: Guggenheim B (ed). Periodontology Today. Int Congr, Zürich. Karger: Basel; 1988;59-67.Listgarten MA, Helldén L. Relative distribution of bacteria at clinically healthy and periodontally diseased sites in humans. J Clin Periodontol 1978;5:115-32.Löe H, Anerud A, Boysen H, Morrison E. Natural history of periodontal disease in man. Rapid, moderate and no loss of attachment in Sri Lankan laborers 14 to 46 years of age. J Clin Periodontol 1986;13:431-40.Löe H, Ânerud À, Boysen H, Smith M. The natural history of periodontal disease in man. Study design and baseline data. J Periodontal Res 1978;13:550-62.Löe H, Theilade E, Jensen SB. Experimental gingivitis in man. J Periodontol 1965;36:177-87.Loesche WJ. Chemotherapy of dental plaque infections. Oral Sci Rev 1976;9:65-107. Löst C. Depth of alveolar bone dehiscences in relation to gingival recessions. J Clin Periodontol 1984;11:583-9.Macgregor IDM. Raidographic survey of periodontal disease in 264 adolescent school­boys in Lagos, Nigeria. Community Dent Oral Epidemiol 1980;8:56-60.Maiden MFJ, Carman RJ, Curtis MA, Gillett IR, Griffiths GS, Sterne JAC et al. Detec­tion of high-risk groups and individuals for periodontal diseases: laboratory markers based on the microbiological analysis of subgingival plaque. J Clin Periodontol 1990;17:1-13.Mandell RL. A longitudinal microbiological investigation of Actinobacillus£ actinomycetemcomitans and Eikenella corrodens in juvenile periodontitis. Infect Immun 1984;45:778-80.Mann J, Cormier PP, Green P, Ram CA, Miller MF, Ship II. Loss of periodontal attachment in adolescents. Community Dent Oral Epidemiol 1981;9:135-41.Marthaler TM. Discussion: Current status of indices of plaque. J Clin Periodontol 1986;13:379-80.Melnick M, Shields ED, Bixler D. Periodontosis. A phenotypic and genetic analysis. Oral Surg, Oral Med, Oral Pathol 1976;42:32-41.Mierau H-D, Fiebig A. Zur Epidemiologie der Gingivarezessionen und möglicher klini­scher Begleiterscheinungen. Untersuchungen an 2410 18-22jährigen (1. Mitteilung). Dtsch Zahnärztl Z 1986;41:640-4.Miller AJ, Brunelle JA, Carlos JP, Brown LJ, Löe H. Oral health of United States adults. The National Survey of Oral Health in USA. Employed Adults and Seniors: 1985-1986. National findings. NIH Pubi 1987 Aug. Report No. 87-2868.Modéer T, Odenrick L. Post-treatment periodontal status of labially erupted maxillary

47

canines. Acta Odontol Scand 1980;38:253-6.Newman MG, Socransky SS. Predominant cultivable microbiota in periodontosis. J Periodontal Res 1977;12:120-8.Newman MG, Socransky SS, Savitt ED, Propos DA, Crawford A. Studies of the microbiology of periodontosis. J Periodontol 1976;47:373-9.Offenbacher S, Weathers DR. Effects of smokeless tobacco on the periodontal mucosal and caries status of adolescent males. J Oral Pathol 1985;14:169-81.O'Lleary TJ, Drake RB, Crump PP, Allen MF. The incidence of recession in young males: A further study. J Periodontol 1971;42:264-7.Paloheimo L, AinamaJ, Niemi M-L, Viikinkoski M. Prevalence of and factors related to gingival recession in Finnish 15- to 20-year old subjects. Community Dent Health 1987;4:425-36.Papapanou PN, Wennström JL, Gröndahl K. Periodontal status in relation to age and tooth type. A cross-sectional radiographic study. J Clin Periodontol 1988;15:469-78. Paterson JR. The etiology of gingival recession. Review of literature. J Ind Dent Assoc 1979;58:33-7.Perry DA, Newman MG. Occurrence of periodontitis in an urban adolescent population. J Periodontol 1990;61:185-8.Persson LÀ. Nutrition and health in infancy and childhood. An epidemiological ap­proach to the assessment of dietary habits, their determinants and implications (dissertation) Umeå: Univ of Umeå 1984;New Series No. 119.Pilot T, Barmes DE, Leclercq MH, McCombie BJ, Sardo Infirri J. Periodontal condi­tions in adults, 35-44 years of age: an overview of CPITN data in the WHO Global Oral Data Bank. Community Dent Oral Epidemiol 1986;14:310-2.Poison AM, Caton JG. Current status of bleeding in the diagnosis of periodontal dis­eases. J Periodontol 1985;56:Suppl 11:1-3.Ralls SA, Cohen ME. Problems in identifying "bursts" of periodontal attachment loss. J Periodontol 1986;57:746-52.Ramjjord SP. Indices for prevalence and incidence of periodontal disease. J Periodontol 1959;30:51-9.Russell AL. A system of classification and scoring for prevalence surveys of periodon­tal disease. J Dent Res 1956;35:350-9.Russell AL. The prevalence of periodontal disease in different populations during the circumpubertal period. J Periodontol 1971;42:508-12.Sangnes G, Gjermo P. Prevalence of oral soft and hard tissue lesions related to mechanical toothcleansing procedures. Community Dent Oral Epidemiol 1976;4:77-83. Saxby M. Prevalence of juvenile periodontitis in a British school population. Commu­nity Dent Oral Epidemiol 1984;12:185-7.Saxby MS. Juvenile periodontitis: an epidemiological study in the west Midlands of the United Kingdom. J Clin Periodontol 1987;14:594-8.

48

Saxén L. Prevalence of juvenile periodontitis in Finland. J Clin Periodontol 1980a;7:177-86.Saxén L. Heredity of juvenile periodontitis. J Clin Periodontol 1980b ;7:276-88. Schenkein H, dan do la L, Genco RJ. Complement activation in gingival pocket fluid from patients with periodontosis and severe periodontitis. IADR J Dent Res 1976;Abstr 581:B207.Schoo WH, van der Velden U. Marginal soft tissue recessions with and without attached gingiva. A five year longitudinal study. J Periodontal Res 1985;20:209-11.Schiirch Jr E, Minder CE, Lang NP, Geering AH. Periodontal conditions in a randomly selected population in Switzerland. Community Dent Oral Epidemiol 1988;16:181-6. Sewón L, Parvinen T. The prevalence of periodontal bone loss in Finnish adults measu­red using simplified radiographic criteria. Proc Finn Dent Soc 1988:84:79-83.Sheiham A. Screening for periodontal disease. J Clin Periodontol 1978;5:237-45. Skaleric U, Kovac-Kavcic M. Periodontal treatment needs in a population of Ljubljana, Yugoslavia. Community Dent Oral Epidemiol 1989;17:304-6.Slots J. The predominant cultivable organisms in juvenile periodontitis. Scand J Dent Res 1976;84:1-10.Slots J. The predominant cultivable microflora of advanced periodontitis. Scand J Dent Res 1977a;85:114-21.Slots J. Microflora in the healthy gingival sulcus in man. Scand J Dent Res 1977b:85:247-54.Slots J, Möenbo D, Langebaek J, Frandsen A. Microbiota of gingivitis in man. Scand J Dent Res 1978;86:174-81.Socransky SS, Haffajee AD, Goodson JM, Lindhe J. New concepts of destructive periodontal disease. J Clin Periodontol 1984;11:21-32.Sterne JAC, Curtis MA, Gillett IR, Griffiths GS, Maiden MFJ, Wilton JMA et al. Statistical models for data from periodontal research. J Clin Periodontol 1990; 17:129- 37.Suzuki JB. Diagnosis and classification of the periodontal diseases. Dent Clin North Am 1988;32:195-216.Suzuki JB, Collison BC, Falkler Jr WA, Nauman RK. Immunologic profile of juvenile periodontitis. II. Neutrophil chemotaxis, phagocytosis and spore germination. J Periodontol 1984a;55:461-7.Suzuki JB, Park SK, Falkler Jr WA. Immunologic profile of juvenile periodontitis. I. Lymphocyte blastogenesis and the autologous mixed lymphocyte response. J Periodon­tol 1984b;55:453-60.Tenenbaum H. A clinical study comparing the width of attached gingiva and the preval­ence of gingival recessions. J Clin Periodontol 1982;9:86-92.Topic B. Classification of periodontal diseases. Int Dent J 1990;40:171-5. van der Velden U, Abbas F, van Stenbergen TJM, de Zoete OJ, Hesse M, de Ruyter C et al. Prevalence of periodontal breakdown in adolescents and presence of Actinobacil- lus actinomycetemcomitans in subjects with attachment loss. J Periodontol 1989;60:604- 10.

van Dyke TE, Zinney W, Winkel K, Taufiq A, Offenbacher S, Arnold RR. Neutrophil function in localized juvenile periodontitis. Phagocytosis, superoxide production and specific granule release. J Periodontol 1986;57:703-8.

49

Vehkalahti M, Helminen S, Rytömaa I. Caries decline from 1976 to 1986 among 15- year-olds in Helsinki. Caries Res 1990;24:279-85.Vrbic V, Vulovic M, Rajic Z, Topic B, Tatic E, Malic M et al. Oral health in SFR Yugoslavia in 1986. Community Dent Oral Epidemiol 1987;16:286-8.Watson PJC. Gingival recession. J Dent 1984;12:29-35.Wennström JL. Lack of association between width of attached gingiva and development of soft tissue recession. A 5-year- longitudinal study. J Clin Periodontol 1987;14:181-4. Wilson JMG, Jungner G. Principles and practice of screening for disease. Public Health Papers no. 34. Geneva:WHO; 1968.Wilton JMA, Curtis MA, Gillett IR, Griffiths GS, Maiden MFJ, Sterne JAC et al. Detec­tion of high-risk groups and individuals for periodontal diseases: laboratory markers from analysis of saliva. J Clin Periodontol 1989;16:475-83.Wilton JMA, Griffiths GS, Curtis MA, Maiden MFJ, Gillett IR, Wilson DT et al. Detec­tion of high-risk groups and individuals for periodontal diseases. Systemic predisposi­tion and markers of general health. J Clin Periodontol 1988;15:339-46.Wolfe MD, Carlos JP. Periodontal disease in adolescents: epidemiologic findings in Navajo Indians. Community Dent Oral Epidemiol 1987;15:33-40.Wolf son SM, Lewis MH. A survey of periodontal conditions in Saskatchewan adole­scents. J Canad Dent Ass 1985;51:486-9.Woofter C. The prevalence and etiology of gingival recession. Periodontal Abstracts 1969;17:45-50.World Health Organization. Periodontal disease. Report of an expert Committee on Dental Health. Geneva: World Health Org Techn Rep Ser; 1961 ;No. 207:1-42. Yoneyama T, Okamoto H, Lindhe J, Socransky SS, Haffajee AD. Probing depth, attach­ment loss and gingival recession. Findings from a clinical examination in Ushiku, Japan. J Clin Periodontol 1988;15:581-91.Zambon JJ. Actinobacillus actinomycetemcomitans in human periodontal disease. J Clin Periodontol 1985;12:1-20.

50