J Clin Feriodoniol I99S: 25: 297-305 Copyright © Munksgaard I<^<JH

Printed in Denmark . .All rigiits n-.wrved

JIUMUlOf
Clinical periodcntology
/.S.V,V OJ

Relationship between smoking p. Axelsson' 2, J.

and J. Lindhe^
'Department of Periodontology. Faculty of
Odontology, Goteborg University, Sweden;

and dental status in 35-, 50-, 65-, ^Department of Preventive and Community
Dentistry, Public Dental Health Service,
Karlstad, Varmland. Sweden

and 75-year-old individuals

A.xels.san P. PauUtiuler J. Limlhe J: Relatiomhip between smoking and dental .statu.s
in 35-. 50-, 65-, ami 75-year-okl individuals. J Clin Periodontol 1998: 25: 297-305.
© Munksgaard. 1998.

Abstract. The aim ofthe present study was to examine the dental status and
smoking habits in randomized samples of 35-, 50-, 65-, and 75-year-old subjects
(n=1093), recruited tor a cross-sectional epidemiological study in the County of
Varmland. Sweden. The following clinical variables were recorded by 4 well-
calibrated dentists: number of edentuolous subjects, number ol' missing teeth,
probing attachment level, furcation involvement, CPITN scores, DMF surfaces,
plaque and .stimulated salivary secretion rate (SSSR). In addition, the subjects
reported in a questionnaire their tobacco habits, orai hygiene habits, dietary
habits etc. The percentage of smokers in 35-, 50-, 65-. and 75-year-olds was 35%,
35'K), 24% and 12%, respectively. In 75-year-olds, 41"/) ofthe smokers were eden-
tulous compared to i5"/« of non-smokers. The difference in number of missing
teeth between smokers and non-smokers was 0.6 (p^0.\5). 1.5 (;?^O.OI3). 3.5 {p^
0.00Q7) and 5.8 (/?^0.005) in the 4 age groups. Smokers had the largest mean
probing attachment loss in all age groups. The ditTerences between smokers and
non-smokers in mean attachment level were 0.37 (/j^O.OOl), 0.88 (;)=O.OOI), 0.85
(p-0.001) and 1.33 mm ( ; J - 0 . 0 0 3 ) in the 35-. 50-, 65-, and 75-year-olds, respec-
tively. Treatment need assessed by CPITN was in al! age groups greatest among
smokers. The number of intact tooth surfaces was fewer in 35-, 50-, and 75-year-
old smokers than in non-smokers. The number of missing surfaces (MS) was
higher in 50-, 65-. and 75-year-old smokers than in non-smokers. In addition. 35-
year-old smokers exhibited a significantly larger number of decayed and filled
tooth surfaces (DFS) than non-smokers. Male smokers had significantly higher
SSSR than non-smoking males (;?=0.012). Plaque index and oral hygiene were
similar in smokers and non-smokers. Smokers reported a more frequent intake of
sugar containing soft drinks (yj=O.OOO) and snacks (/f=0.003) than non-smokers. Key words: analytic epidemiology; smoking;
tooth loss: attachment loss; CPITN; caries
The opposite was reported for consumption of fruit (/;^0.003). It was con- prevalence; plaque; salivary secretion rate;
cluded that smoking is a significant risk indicator for tooth loss, probing attach- oral hygiene habits; dietary habits
ment loss and dental caries.
Accepted for publication 21 July 1997

The main objective of epidemiological
surveys of a cross section design is to
provide important information about
the distribution of e.g. oral diseases in
a population, but the data generated
can also be used to assess the relative
importance of indicators for a given
disease or group of diseases (Beck et
al. 1990, Hansen et al. 1990, Horning
et al. 1992, Grossi et al. 1994. Beck
1994, Wolff et al. 1994, Wiktorsson
1995).
Findings from a number of cross-sec-
tional studies have indicated that
smokers have more missing teeth and
more advanced periodontal disease
than non-smokers (Table 1). Most of
the studies referred to were carried out
in selected groups of adults stratified
into age intervals. Such a selection may
result in a scewed distribution of indi-
viduals recruited to the study, and this
may influence the interpretation of the
data obtained. To our knowledge, no
study has been published in which the
effect of smoking on both caries and
periodontal diseases was evaluated in a
randomized sample of adults belonging
to well defined age groups (indicator
age groups).
The aim ol" the present study was to
assess the prevalence and severity of
periodontal disease and dental caries in
smokers and non-smokers in a ran-
domized sample of 35-, 50-. 65-. and 75-
year-old subjects.
298 A.xels.';on et al.

Material and Methods
An epidemiological oral survey was
perlbrtned on 35-, 50-, 65-. and 75-
year-old subjects in the Coiinty of
Varmland, Sweden. The age groups
were selected according to the recom-
mendations of the Swedish National
Board of Health and Welfare (Sund-
berg 1989). A randomized, stratified
sample of 35-yecir-olds (/j=155), 50-
year-olds (H^510), 65-year-olds («=
310) and 75-year-olds (/j-310) was
drawn from approximately 220,000 in-
habitants in the county. Stratification
was based on both age and urban/ru-
ral living - 5O'!^p of the subjects lived
in an urban area (>50,000 inhabi-
tants) and 50% lived in 4 rural areas
(<8,000 inhabitants in the major mu-
nicipality). It should be observed that
Ihe number of 50 year old individuals
included in the present study is con-
siderably larger than the number of
subjects in the 3 remaining age cate-
gories. The reason for this is that the
50 year old subjects were intended to
be included also in a 10 year longi-
tudinal epidemiological survey.
In the 35-, 50-. 65-. and 75-year age
categories 11, 64. 45 and 72 subjects, re-
spectively, failed to appear for the
examination. The main reasons for not
attending were (!) "no longer living at
the official address" (tnostly 50-year-
olds), (2) "illness", particularly among
65-, and 75- year-olds and (3) "refusal
to participate".
Before the clinical examination, all
subjects signed informed, written con-
sent regarding participation. Individ-
uals reporting "no smoking" habits and
"smoking" habits were selected for the
present study while "former smokers"
were excluded. The resulting distri-

Table 1. Review of cross-sectional orai epidemiology studies related to smoking habits

Materials Variables
Periodontal disease Life style
Studies n Age SEL RAN AG MT AB VAL HAL CPITN GI PD CAR PI PP CP SSR DC OH DH OTH
Herulf (I95II 535 19-25 X mt
Massler
et al. (1952)2577 17-21 mt NS
Arno et al. (195K)1246 25-55 mt S
Arno et al. (19591 728 21-45 mt S
Heruir (19681 700 20-85 int s s
Bergstrom
et al, (1983) 328 39-78 X int S s -s NS
Feldman
et ai. (1983) 862 7 all s NS NS -s
Preber et al. (1986) 369 18 70 int -s NS NS
Osterberg
et a!. (1986)1377 70 X ind
Bergstrom
et al. (1987) 235 21-60 int NS
Beck et al. int
Hansen (1990) 690 65 +
et al. ind NS
Goultscliin (1990) 144 35
et al. int
Bergstrom (1990) 344 20-70
et al. all
Horning (1991) 210 24-60
et al. 13 84 all
Preber et al. (1992)1783 32 74 all NS
Stoltenberg (1992) 145
et al, (1993) 189 28-73 all NS S NS
Haber et al. (1993) 227 int S
Linden et al. (1994) 82 20-33 all S S -S NS NS
Grossi et al. (1994)1426 25-74 all
Martine?-
Canut et al. (1995) 889 21-76 all
Wictorsson (1995) 236 44-56 all NS
Sakki et al. (1995) 527 55 ind S
Soder et al. (1995) 144
s
all s s NS S NS
Mtillally
et al. (1996) 100 35 + all NS NS
SELT: study on selected material; RAN: study on randomized material; .4G: agegroup (statistics performed on: ini: age interval; ind: indicator
age group(s); all: all ages simultaneously): MT: number of missing teeth; AB: alveolar bone; VAL: vertical attachment level: HAL: horizontal
attachment level; CPIT!^: community periodontal index of treatment needs; GI: gingival index; PD: pocket depth; CAR: dental caries; PI:
plaque index; PP: periopathogens; CP: caries pathogens; SSR: stimulated saliva secretion rate; DC: dental care habits; OH: oral hygiene
habits; DH: dietary habits; OTH: other important features. NS: non significant difference; S: significant difference, non smokers beeing
healthier; —S: reversed significant difference.
 Smoking and dental health .status 299

Table 2. The primary sample, former smokers, smokers and non smokers according lo age and gender
Age (years) 35 50 65 75 TOT
Primary n 11 n
sample 155 510 310 310 1285
Examined 144 446 265 238 1093
male I "M female ("•''<•) m a l e ( "'il) female {"Al) male ( %) female {"/ '•<) male ( "/•i) female ("A,) male i1*!'") female ("/,)
72 72 215 231 128 137 117 121 536 557
Smoking
unknown 0 (0) 1 (l.4| 1 (0.5) 0 (0) 4 (3.1) I (0.7) 6 (5.0) 8 (6.8) ll (2.1) 10 {1.8)
Former
smokers 7 (9.7) 9 (12.5) 36 (6.7) 22 (9.5) 15 (11.7) 9 (6.6) 26 (21.5) 7 (6.0) 84 (15.7) 47 (8.4)
Smokers 21 (29.2) 23 (31.9) 65 (30.2) 71 (30.7) 37 (28.9) 19 (13.9) 14 (11.6) 8 (6.8) 137 (25.6) !21 (21.7)
Non-
smokers 44 (61.1) 39 (54-2) 113 (52.6) 138 (59.7) 72 (56.3) 108 (78.8) 75 (62.0) 94 (80.3) 304 (56.7) 379 (68.0)

bution of subjects in the diita base is
shown in Table 2.
Oral examination
The examiniilion was performed in a
public dental health clinic, using a con-
ventional dental unit and illumination.
Contamination with saliva was pre-
vented by the use of saliva ejectors and
cotton roils. The teeth were dried with
compressed air and examined wiih the
aid of dental mirrors and explorers. A
set of full mouth radiographs was taken
in the 50-, 65-, and 75-year-olds, and
bite-wings were taken in the 35-year-
olds.
Variables
The following variables were recorded:
(1) number of teeth, (2) probing attach-
ment level. (3) furcation involvement.
(4) CPITN, (5) prevalence of dental
caries and fillings. (6) plaque index (PU)
and (7) stimulated salivary secretion
rate (SSSR). Each participant com-
pleted a questionnaire on ongoing or
former tobacco use, "dry mouth", oral
hygiene and dietary habits.
Number of teeth: roots with lack of
crown substance were reported as ab-
sent. Third molars were excluded.
Probing attachment level: measure-
ments were performed aecording to
Ramfjord (1959) using a double-ended
WHO probe. All measurements were
made in millimetres. All mesial surfaces
were measured.
Furcation involvement score 0, 1 and
missing: Only molars were evaluated.
The measurements were made at each
furcation entrance using a double-end-
ed curette (Goldman-Fox #3). Score 0
was recorded when the curette did not lets), was scored on all mesial, buccal,
"catch" at the furcation entrance and distal and lingual tooth surfaces in 50-
score 1 as soon as the tip of the curette year old subjects. The % of surfaces
reached the furcation area by gentle with plaque was calculated.
probing. Thus the furcations were Stimulated saliva secret ioti rate
dichotomized into healthy (score-0) iSSSRj: the volume (ml) of saliva,
and involved (score=l). For extracted sampled per minute and produced by
teeth, the furcations were recorded as paraflin chewing was measured in 50-.
missing. 65-. and 75-year old subjects.
Periodontal trealment needs: meas- Frequeney of orai hygiene measures:
urements were made according to the ihe use oi' toothbrush, dental Hoss.
Community Periodontal Index of Treat- interdental toothpicks and interdental
ment Needs - CPITN - (Ainamo 1984). brushes was reported on an ordinal as-
All mesial, buccal. distal and lingual cending scale from 1-"never" to 7 -
surfaces were defined according to score "every day, morning, evening and after
0, 1, 2. 3. 4 or missing. every meal".
Caries prevalence: clinically visible Frequeney of jood consumption: the
primary and secondary (recurrent) actual number of intakes was recorded.
caries lesions, radiographic caries and The intake of sugar containing items
filled surfaces were registered in accord- (sweets, confectionary, soft drinks,
ance with Axelsson & Lindhe (1978). jam), snacks and fruit was reported on
Clinical examination was carried out an ordinal ascending scale from 1 =
using a WHO double-ended probe. "never" to 5 = "more than seven times
Dental bitewing radiographs, 2 or 4, de- per week".
pending on the width ofthe dental arch, Feeling of "dry mouth": was recorded
were used to detect decayed and filled as l^"never", 2^"seldom", 3^"often"
approximal surfaces. The number of de- and 4="always".
cayed-, missitig- and filled (DMF) teeth The data from the clinical examin-
and surfaces was calculated. ation were transferred to a specially de-
Plaque (PIl): the presence of plaque signed computer program. Data rel-
(after disclosure with erythrosin pel- evant for the studv were retrieved, re-
Tahle 3. Mean number of missing teeth according lo ;ige and smoking habits; dentate subjects
only (/( = 808)
Age (years) 35 50 65 75
Smoking I + +
/; 83 44 242 127 145 45 109 13
mean 1.4 2.0 4. 8 6.3 10.3 13.8 13.0 18.8
SD 1.82 1.98 5. 13 5.44 6,51 7.63 6.89 6.44
difference -0. 6 -1.5 -3.5 -5.8
p=0. 15 /T=0.013 /) = 0.007 p=0.005
95 % CI -1.343; 0.207 - 2 . 570; -0.311 -6.026: -0.995 -9.772; -1.821
300 A.xels.son et al.

modelled and transferred into ti
statistical database package (SPSS^)
together with answers from the qttes-
tionnaire.
4 teams of dctital examiners were
trained and calibrated. An ititer- and
intra- examiner reliability study was
performed on 20. 35-50-year-old sub-
jects, and 20 subjects 65-75-years old.
Each of the 40 subjects was examined ^
times. The reliability study included the
foliowing variables: caries prevalence
(DFS). probing attachment level (tnm)
and furcation involvement (score 0 1).
The outcome of the reliability study in-
dicated that the 4 examiners could be
used interchangeably for: caries preva-
lence (intra class correlation coef-
tieient=0.996/().996 and 0.980/0.990;
variance due to examiners O.l/O.OI'Mi
and 0.06/0.02'/<) for total and approxi-
ma! DFS in 35-50- and 65-75-year-
olds, respectively) and probing atlach-
tnent level tiieasiirenient (intra-class
correlation coefficient 0.924 and 0.933;
variance due to examiners l.3/3.O'/<p in
35-50 and 65-75 year olds, respec-
tively). For the assessement of furcation
involvetnent, however, the level of
agreement was low (intra class corre-
lation coefficient=0.682 and 0.745;
variation due to difference between the
examiners was not significant and vari-
ance due to examiners in relation to
total variance was negligible in both age
groups).
Statistics
Analysis was performed on a personal
computer using the SPSS® system.
Non-parametric variables were tested

100 100

50 -

100 100
17/47 15/45 13/43 11/41 22/32 24/34 26/36 17/47 21,5 13/43 11/41 22/32 24/34 26/36
16/46 14/44 12/42 21/31 23/33 25/35 27/37 16/46 14/44 12/42 21/31 23/33 25/35 27/37
smokers non-smokers smokers non-smokers

100 100

50

100 100
17/47 15/25 13/43 11/41 22/32 24/34 26/36 17/47 15/25 13/43 11/41 22/32 24/34 26/36
16/46 14/44 12/42 21/31 23/33 25/35 27/37 16/46 14/44 12/42 21/31 23/33 25/35 27/37
u ^ smokers
smoKers non-smokers
non-smoKers a • smokers
smokers non-smokers
Fig. L Percentage of missing teeih. (a) 35-year-olds; (b) 50-year-olds; (c) 65-year-olds; (d) 75-year-olds
 Smoking and dental hccilth .status 301

mm jects, there was only a minor difference

35 yrs 50 yrs
~ non-smokers
Fig. 2. Mean mesial loss of probing attachment.
using the Mann-Whitney U - Wilcoxon
rank sum test and the y~ test. Para-
metric tests were performed by the Stu-
dent /-test. Although some data are rep-
resented in the Tables and Figures as
'^s. all statistical calculations were
based on raw data.
Results
The overall frequency of smokers was
27.4%. Approximately 32-37'^ of the
35-50- and 65-year-old males were
smokers but only 15,7"'ti of the 75-year
olds. In females, the highest frequency
of smokers was found among Ihe 35-
year-olds (37,r'/i) and the lowest in the
75-year olds (7.8'KJ). Significantly more
males than females in the 65-vear-old
65 yrs 75 yrs
• smokers
category were smokers 33.9'^ versus
15.0% (77=0.0006).
There were no edentulous subjects
among the 35-year-olds, The '/^J of eden-
tulou.s subjects in non-smokers versus
smokers increased with age from 3.6/
6.6. 19,4/19,6 to 35.5/40,9% in 50-, 65-.
and 75-year-oids. but there were no
statistically significant differences be-
tween groups.
Tooth loss was on the average greater
in smokers than in non-smokers. The
differences regarding the number of
missing teeth were statistically signifi-
cant in the 50-, 65-, and 75-year age
groups(/?-0.013,/»-0.007./)-0.005),as
reported in Table 3. The distribution of
missing teeth in the dentition ("pattern")
is presented in Fig. I. In 35-year-old sub-
in the "pattern" of missing teeth between
smokers and non-smokers: except for
the first molars, which were more often
missing in smokers. The 65- and 75-year-
old age category smokers exhibited a
much higher percentage of missing
maxillary anterior teeth than non-
smokers, but in both groups the molars
were the most frequently missing teeth.
The mean probing attachment level at
mesial surfaces increased with age in all
age categories. Smokers displayed a
higher probing attachment loss (PAL)
value than non-smokers. The differences
regarding PAL between smokers and
non-smokers (0.37, 0.88, 0.85 and L33
mm) in the 4 age groups (Fig, 2) were
statistically significant (/t=0,001. p=
0.0000. /j-0,0000 and /)-0,002) in the
35-. 50-. 65-, and 75-year-olds respec-
tively.
The "/<• of molars with furcation in-
volvement was in all age groups greater
in smokers than in non-smokers, and the
difference increased with age; 3.6, 13.8,
19.3 and 26.5 % for 35-. 50-, 65-. and 75-
ycar olds respectively. The / - test showed
these differences to be statistically sig-
nificant in 3 age groups: 35-. 50-, and 65-
year old subjects (Table 4).
The mean "/.s of different CPITN
scores is reported in Fig, 3, The % of
score 0 in non-smokers versus smokers
was 71/54. 55/43. 37/27 and 30/17 and
for missing surfaces 5/7, 18/23, 37/49
and 46/67 in the 35-. 50-. 65-. and 75-
year-olds, respectively. In all age
groups, the "A> of sites with CPITN
score 1 was lower in smokers than in
non-smokers. The differences between
smokers and non-smokers regarding the
mean number of missing surfaces were
2.2 (/;-O.I58), 5.8 (/?-0?013). 14.0 (p=
0.0007) and 23.1 (/?-0.005) in the 35-.

Table 4. Dentate subjects (« = 8O8}; number of molars uiih iLirciition involvement related to total number of molars; dilTerence in mean number
of molars with furcation involvement related lo smoking habits. (/; = 679}
35 50 65 75
Age (years) ("•") (%) ( (%)
Smoking - 4- - + - + - +
subjects without molars 0(0) 0(0) 16(7) 8(6) 35 (25) 15 (33) 43 (39) 11 (85)
subjects with molars 83 (100) 44(100) 226 (93) 119(94) 109 (75) 30 (67) 66(61) 2(15)
total number of molars 636 314 1287 590 395 88 203 5
mean 7,7 7.1 5.7 5,0 3,6 29 3,1 2,5
SD 0,887 1,424 2.066 2.164 2,004 2 116 1,739 0.707
furcation involved molars 17(2,7) 20 (6,3) 186(14.5) 167(28,3) 88 (22,3) 37 (42,0) 68 (33.5) 3 (60.0)
not furcation involved
molars 6!9(97,3) 294 (95,7) itOl (85.5) 423(71.7) 307 (77.7) 51 (68.0) 135(66,5) 2 (40.0)
/ - test 7.67 50.84 14.66 1.52
P /XO.OOl /XO.OOI
302 Axelsson et al.

The mean number and the differ-

ences in Intact, Decayed. Filled, and
Missing surfaces in dentate subjects
were calculated (Table 5). The 35-,5Q-.
30 and 75-year-old non-smokers had sig-
nificantly larger numbers of intact sur-
faces than the smokers. Significantly
more filled surfaces were recorded in 35-
year-old smokers than in non-smokers,
but the 65-year-old non-smokers had
more filled surfaces than smokers.
Smokers had significantly more missing
surfaces than non-smokers in the 50-,
65-. and 75-year-old age groups. In the
50-year-olds, there was no significant
difference between non-smokers and
smokers with respect to the % ' > of sur-
faces with plaque.
•10 The SSSR in a combined group of
50-, 65-, and 75-year-old males and fe-
males was calculated. Male smokers
and non-smokers had a mean secretion
rate of 2.1 and 1.8 ml/min. respectively.
so
The difference, 0.3 ml/min, was statisti-
cally significant (p=O.OI2, 95% CI
0.575; 0.073). For females, there was no
statistically significant difference in sali-
35 yrs 50 yrs B5 yrs 75 yrs vary secretion rate between smokers
l ^ l missing | I score 0 Y//A score 1 ^ ^ score 2 l_ ! score 3 ^ H score 4 and non-smokers. Smokers more fre-
quently reported a feeling of "dry
Fig. i. Percentage of CPITN scores and missing surfaces. "-"" refers to non-smokers and "• + " mouth" (/.'^0.004). There were no stat-
to smokers.
istically significant differences in the
frequency of using different oral hy-
giene measures (toothbrush, inlerdental
floss, interdental toothpicks and inter-
50-, 65-. and 75-year-o!d age groups. In ( —). Statistically significant differences
dental brushes) between smokers and
35-year-olds, there were stalislically sig- in the 50-year-olds were found for
non-smokers. Concerning dietary
nificant differences between non- CPITN score 0 (+), 3 ( - ) and 4 ( - ) .
habits there were no significant differ-
smokers and smokers in the mean num- In the 65- and 75-year-oIds, statistically
ences between smokers and non-
ber of CPITN score 0 - positive differ- significant differences were found for
smokers in the mean number oi' meals
ence ( + ). 2 and 3 negative differences CPITN score 0 (+) and 1 ( + ).

Table 5. Mean number of inlacl. decayed, filled and missing surfaces in dentate subjects (« = 808)
35 50 65 75
Age (years) *(=127 n=369 n=[90 n = 122
smoking - + + + +
n 83 44 242 127 145 45 109 13
mean inlacI 89.9 79. t 51.3 43.6 35.0 29.2 27.8 13.4
SD 1S.91 18.66 21.50 22.71 23.27 19.48 22.12 13.74
mean decayed 0.3 1.5 0.4 0.6 0.2 0.5 0.2 0.15
SD 0.61 5.68 1.90 2.00 0.86 f.O4 0.71 0.56
mean filled 30.9 37.8 52.7 53.4 43.5 33.1 38.2 26.5
SD 16.50 14.65 18.77 21.27 20.68 24.52 22.30 19.40
mean missing 6.9 9.7 23.6 30.4 49.3 65.2 61-8 87.9
SD 8.87 12.28 24.20 25.64 30.14 34.79 31.51 28.75
diff intact 10.8/7=0.002 7.7/1=0.002 5.877=0.130 14.5/7=0.003
95% CI 3.90f; f7.794 2.944; 12.391 -1.734; 13.372 5.375; 23.544
diff decayed -I.2/J=O.183 -O.\ p=0.5\6 -0.3/7=0.122 0.07 /7=0.746
95% CI -2.894; 0.569 -0.554; 0.279 -0.606; 0.073 -0.339; 0.472
diff filled -6.9/-=0-02l -0.7/7=0.732 10.4/7=0.012 11.6/7=0.075
95% CI -12.769: -1.042 -4.975; 3.498 2.344; 18.474 -1.172; 24.425
difi" missing -2.8/j=0.145 -6.8/7=0.013 -16.0/7=0.007 -26.2/>=0.005
95"^ C! -6.534; 0.975 -12.ff5; - L 4 6 8 -27,459; -4.464 -44.304; -8.001
 Smoking and dental health status 303

(4.1/4.0). intake of sweets or sugar con- in a longitudinal study (3 years) showed a case control study in a selected group
taining confectionary. Sugar containing that smokers (>50-years-old) lost 0.8 of referred periodontal patients, re-
soft drinks and snacks were consumed teeth / individual, compared to 0.4 lost ported that smokers had on average 1.9
more often by smokers than by non- teeth in non-smokers {p<0.05). molars with furcation involvement com-
smokers (/j^O.OOO and / J = 0 . 0 0 3 respec- The 65- and 75-year-o!d smokers pared to 0.9 in non-smokers.
tively) whereas fruit intake was more (Fig. lc. d) had a higher % of missing In all age groups, the "/.>s, of healthy
frequent in non-smokers (/J=0.003). maxillary incisors than non-smokers. sites (CPITN score^O) were lower in
The 75-year-old smokers had also more smokers than in non-smokers, but the
missing maxillary premolars and mo- "Al of sites with gingivitis and shallow
Discussion lars than non-smokers (Fig. Id). This pockets (CPITN score- 1) was higher in
Approximately 30% of the subjects re- may indicate that smoking has a local non-smokers than in smokers. These
cruited for the study were smokers, but deleterious effect in the maxillary den- findings are in agreement with data
the distribution was tinequal with re- tition and in particularly on the front from other studies showing that
spect to age group and gender; 37% of tooth region. In all four age groups smokers experience less gingivitis than
35-year-old females were smokers, but smokers exhibited significantly more non-smokers (Bergstrom & Floderus-
only 8'/o of 75-year-olds. TKe corre- probing attachment loss than non- Myrhed 1983, Preber & Bergstrom
sponding values for mates were 'i2% smokers (Fig. 2). This difference in PAL 1985. 1986. Bergstrom 1990, Gouldtsh-
and 16'/o respectively. Overall, the distri- increased with age. Thus the present re- in et al. 1990). One reason for lower
bution of smokers related to age and sults indicated that loss of periodontal gingivitis scores in smokers than in
gender was found to be in agreement attachment was correlated with the non-smokers is that smoking appears to
with data for Sweden previously re- number of years the subjects had been suppress the vascular reaction in the
ported (Swedish National Board of smokers. The current findings are thus marginal gingiva. In a study by Dani-
Health and Wellfare Report 1991). in agreement with data from other elsen et al. (1990) the rate of plaque ac-
cross-sectional studies showing that cumulation was similar in smokers and
In the current study, the "A> of edentu- smoking is a powerful risk indicator tor non-smokers, but the increase in gingi-
lous subjects among non-smokers and loss of periodontal tissue support (Beck val vascuiarity in smokers was only half
smokers was similar in the four age et al. 1990. Linden & Muilally 1994. of that in non-smokers. The proportion
groups. In 75-year-old males, however. Grossi et al. 1994. Martinez-Canut et of CPITN scores i-4- among the re-
43'yii of the smokers were edentulous, al. 1995). maining sites in the present material,
compared to 35"/i of non-smokers. This however, was higher in smokers than in
difference between smokers and non- In the baseline data collected for a non-smokers. The findings of the pres-
smokers is in agreement with findings longitudinal epidemiological study on ent study further showed that smokers
by Osterberg & Mellstrom (1986) who older adults (50-64 and 65- 75 years), had significantly fewer intact tooth sur-
found that in 70-year-old males almost Locker & Leake (1993) found that faces in all age groups than non-
50% of smokers were edentulous com- smokers had significantly more loss of smokers, and more missing surfaces
pared to only 20'!^) 0*1 non-smokers. probing attachment than non-smokers. (MS) in 50-. 65-. and 75-year olds. Al-
The difference in the mean number In a recent 3-year prospective study. though it may be argued that most MS
of missing teeth between smokers and Beck et al. (1995) showed that among were lost due to periodonta! disease in
non-smokers increased from the age of older adults, smokers were significantly these age groups, it is interesting to note
35 to 75. This may indicate that the more susceptible to loss of periodontal that in the 35-year olds, smokers had
longer the subjects had been smokers attachment than non-smokers. Machtei significantly more DFS and fewer intact
the more extensive was the relative et al. (1997) also showed in a longitudi- surfaces than non-smokers. This is in
tooth loss. This conclusion is in agree- nal study that smokers had three times agreement with Wictorsson (1995) who
ment with results from a longitudinal more attachment loss and twice more found significantly higher caries preva-
study by Holm (1994). In this study 20 alveolar bone loss than non-smokers. lence among smokers than in non-
40-year-old non-smokers lost, over a 10 Tobacco smoking also seems to play an smokers in a randomized sample of 30-
year period, only 13 teeth per 1000 indi- important role in refractory peri- 40-year old Swedes living in an area
viduals, compared to 121 teeth in age odontitis (Adams 1992, MacFarlane et with low water fluoride concentration.
matched smokers (>15 cigarettes/day). al. 1992). Compared to the % of The above findings indicate that smok-
During the same time interval. 50 70- smokers in the general population ing may indeed also be associated with
year-old smokers lost as many as 543 (<3O'/o) smokers comprised an unusu- increased risk for dental caries. This
teeth per 1000 individuals. Finding by ally high '^ (>90"A<) of refractory peri- hypothesis is supported by finding by
Osterberg & Mellstrom (1986) also con- odontitis cases (MaeFarlane et al. Ravald et al. (1993). In a twelve year
firmed that a correlation exists between 1992). longitudinal study they observed that
the number of lost teeth and the total smokers after periodontal treatment de-
The current data showed that
exposure to smoking. In the Oster- veloped significantly more new root
smokers had a greater prevalence of
berg & Mellstrom material the 70-year- caries lesions than non-smokers.
molars with furcation involvement than
old male smokers had 7 more missing
non-smokers. This difference between
teeth than age matched non-smokers.
the groups increased with age. Al- Plaque was measured only in the ex-
Even in a selected group of young
though no other cross-sectional epide- panded group of 50-year old subjects.
adults (20-35-ycar olds) Linden &
miological studies have compared the No differences in plaque scores were
Mullally (1994) found that smokers had
prevalence of molars with furcation in- observed between smokers and non-
lost 2.2 (p<Q.(i5) teeth more than non-
volvement in smokers and non- smokers. This oberservation is in agree-
smokers. Recently Locker et al. (1996)
smokers, Mullally & Linden (1996), in ment with results from other cross-sec-
304 A.xelssoit et al.
lionai studies (Bcrgslroni & Floderus-
Myrhcd 1983, Preber & Bergstrom
1986. Berg.strom & Eliasson 1987. Lin-
den & Muilally 1994, see Table I). We
found significantly higher SSSR in male
smokers than in non-smoking males
and no differences between female
smokers and non-smokers. In spite of
this smokers more frequently reported
a feeling of ""dry mouth". It has been
proposed (Sheiham 1971. Ismail et al.
1983. Sakki et al. 1995) that the poorer
dental health of smokers may be attri-
buted in part to different life style
habits. In the present study, however, no
differences were found between smokers
and non-smokers with respect to oral
hygiene habits, which are known to in-
fluence dental health status. The mean
number of meals, the consumption of
sweets and sugar containing confection-
ary were similar in smokers and non-
smokers. The frequency of intakes of
sugar containing soft drinks, possibly
due to a feeling of "dry mouth", and
snacks was higher in smokers, which
may result in a higher risk for caries.
In eonelusion. the present study docu-
mented that smoking is a significant
risk indicator for tooth loss, probing
attachment loss and dental caries.
Zusammenfassung
Beziehung zwischen Riiiichen und Zahustatus
bei 35-. 50-. 65-. und 75 Jahrigen Iiulividtten
Das Ziel der vorliegenden Sludic war es. den
Zahnstatus und die Rauciigewohtiheiten in ei-
ner randomisierten Stichprobe von 35-, 50-,
65-, und 75 jiihrigen Individuen ( = 1093) zu
unlersuchen. Die Personen wurden anhand
einer epideniiologischen Querschnitlsstudie
im Bezirk Varmland, Schweden rekruticrt.
Durch vier gut kalibrierle Zahnarzle wurden
die folgendcn klinischen Parameter erhoben:
Anzahl der zahnlosen Personen. Anzalil der
fehlenden Zahne, klinisches Atlachmcntnive-
au, Furkationsbefiill, CPITN-Werte, DMF-
Flachen, Plaque sowie slimuliertc Speichelse-
kretionsrate (SSSR). Zusiilzlich gaben die
Personen in einem Fragebogen ihre Ranch-,
die Mundhygiene- und die EBgewohnhcilen
usw. an. Der Anieil der Raucher bei den 35-,
50-, 65- und 75jahrigen lag bei entsprechend
35%. 35%, 24"/" und 12'^.. Bei den 7?jahrigen
waren A\% der Raucher verglichen mil 35'^!
der Nicliiraucher zahnlos. Die Unlerscliiede
in der fehlendcn Ziihne zwischen Riiuchern
und Nichtrauchcrn bctrugen in den vier
Aitersgruppen 0.6 (/)^O.I5}. 1.5 (/)=0.0I3).
3.5 (/)=0.0007) und S.S (/> = 0.005). Raucher
hatten in alien Aitersgruppen den groUten
miltleren klinischen All;ichmentverlusl. Die
Unterschiede im mitlleren Auachmcniniveau
zwischen Rauchern un Nichtriiuchcrn betru-
gen entsprechend 0.37 (;) = O.OOI), 0.88 (/?=
0.001), 0.85 (;j = O.OOI) und 1.33 mm (;>=
0.002) bei den 35-. 50-. 65-, und 75j;ihrigen.
Der Behandlungsbedarf, gemessen mJltels
CPITN. lag in alien Aitersgruppen bei den
Rauchern am hochstcn. Die Anzahl von in-
laklen Zahnfliicben war bei den 35-, 50-. und
75j;ihrigen Rauchern gerlngcr ais bei den
Niehlrauchern. Die Anzahl der lehlenden
Zahnflachen (MS) war bei den 35-, 50-, und
75jahrigen Rauchern boher als bei den
Nichiranchern. Des weiteren zciglen 35-jahr-
ige Raucher eine signifikant groBere Anzahl
an karioscn und geftillten Zahnfliichen
(DFS) als Nichiraucher. Miinnliche Raucbcr
batten eine signifikatil hohere SSSR als
mannliche Nicbiraucher (/>=0.0]2). Plaque-
Index und Mundhygiene waren bei Rau-
chern und Nichtraiicbern vergleicbbar. Rau-
cher bericliteten iibcr cine hiiufigere Aufnah-
me von zuckcrballigen Erl'rischungsgelran-
ken (77=0.000) und Snacks (/)=0.003) als
Nicliiraucher. Das Gegenteil war bcini Ver-
zchr von Friicblen (/J = 0.003) bcrichtct. Es
wurde die SchluBfolgerung gczogen, daC
Raucben ein signifikantcr Risikoindikalor
fur Zahnverlust, Attachment verlust und
Zahnkaries ist.
Resume
Relation eiiirc le tahac el I'elat ilcniciire ehez
dfs sujets d^es de 35. 50. 65 et 75 ans
Le preseni Iravait avait pour but d'examiner
I'etat dentaire et les liabiiudes eonceinani le
tabac dans des echaniillons randomises dc
sujets iige.s de 35. 50. 65, et 75 ans («= 1 093),
recrutcs pour une elude epidemiologique
Iransversale dans lecomte de Varmlaud, Sue-
de. Les variables eliniques suivantes ont ete
enregistrecs par 4 dentisies calibres pour as-
surer une bonne concordance des resultats:
nombre de sujets edentes, nombre de dents
abscnics, niveau de ['attache au sondage. at-
teinte de la furcation, scores du CPITN (indi-
ce parodontal dc bcsoin en traitements du
public), faces CAO (cariecs, absentes, obtu-
rees) et vitesse de la secretion salivaire stimu-
lee (SSSR). De plus, les sujets onl dans un
questionnaire rendu compte de leurs habitu-
des concernant lc tabac. I'hygiene bucco-den-
taire, I'alimentation, elc. Les proportions de
fumeurs chez les sujets de 35, 50. 65 et 75 ans
etaient respectivement de 35'Mi. 35"/;,, 24"/;i et
12'^ii. Chez les sujets dc 75 ans, AV'Ai des fu-
meurs etaient edentes, a comparer a 35'i^i des
non-fumeurs. Les diflerences entre les nom-
bres dc denis absentes chez les fumeurs et Ies
non-fumeurs etaienl rcspectivement de 0.6
(;!=0.l5), 1.5 (;'=0.013K 3.5 {p^Q.imi) et
5.8 (/7-O.Ofl5) dans les 4 groupes d'agcs.
Dans lous les groupes d"ages, les fumeurs
avaicni la perte d'attacbe au sondage moyen-
ne la plus grande. Les differences entre les
niveaux moycns d'altache chez les fumeurs et
les non-fumeurs etaient respectivement de
0.37 (/) = 0.00l), 0.88 (/J = O.OO1), 0.85 {p =
0.001) ct 1.33 mm/) = 0.002) chez ies sujets dc
35. 50, 65 et 75 ans. Les besoins en traitement
mesure.s par CPITN etaient les plus grands
parmi les fumeurs dans lous les groupes d"a-
ge. Le nombre de faces dentaires intactes
etait plus bas chez les fumeurs de 35, 50, ct
75 ims que chez les non-fumeurs. Le nombre
dc faces dentaires absentes (missing surfaces)
eliiit plus eleve chez les fumeurs de 50, 65 et
75 ans quc chez les non fumeurs. Dc plus, les
fumeurs dc 35 ans presentaient un nombre
significativcment plus grand de faces carices
(decayed) ct obturccs (filled) que les non-t^i-
meurs. Chez les fumeurs du sexc masculin.
SSSR etait significaiivement plus elevee que
chez les non-fumeurs du sexe maseulin (/; =
0.012). L'indice de plaque et I'hygiene bucco-
dentaire etaient semblables chez les fumeurs
et les non-fumeurs. Les fumeurs rendaient
comple d'une consommation plus frequente
de boissons rafraichissantes (/'=0.000) et
snacks (/! = 0.003) conlenant du sucre que les
non-fumeurs. Pour la consommation de frui-
ts, c'etail le contraire (/' = ().003). En conclu-
sion, le tabagisme rcpresente un indicaleur si-
gnificatif du risque dc perte de dents, de per-
le d'attachc au sondage et de caries denlaires.
References
Adams. D. F (iy92) Diagnosis and Ireatmenl
of refractory periodontitis. Current Opin-
ions in Dentistry 2. 33-38.
Ainamo, J. (1984) Assessment of periodontal
treatment needs. Adaptation ofthe WHO
Community I'eriodontal Index of Treat-
ment Needs (CPITN) to European con-
ditions. In A. Frandsen Ed.: Public health
aspects of periodontal tlisect.se in Europe.
Berlin: Quintessenz Vcrlag.
Arno. A., Schei, O,, Lovdal, A. & Waerhaug.
J. (1959) Alveolar bone loss as a function
of tobbaco consumption. Acta Oilontohg-
ica Sctindiiuivica 17. 3-10.
Arno, A.. Waerhaug, J., Lovdal. A. & Schei,
O. (1958) Incidence of gingivitis as related
to sex, occupation, tobacco consumption,
loothbrushing. and age. Oral Surgery Oral
Medicine and Oral Pathology I I . 587-595.
Axelsson, P & Lindhe, J. (1978) Effect of
controlled oral hygiene procedures on
caries and periodontal disease in adults.
Jotirnul of Clinical Pcriodontologv S. 133
151.
Beck, J. D. (19941 Methods of assessing risk
for periodontitis and developing multifac-
toriai models. Journal of Periodontologv.
65, 468^78.
Beck, J. D . Koch, G. G.. & Offenbacher, S.
(1995) Incidence of attacbment loss over 3
years in older adults - new and pro-
gressing lesions. Coimnunily Dentistrv and
Oral Epidemiology 23, 291-296.
Beck, J. D , Koch, G. G., Rozier, R. G. &
Tudor, G. E. (1990) Prevalence and risk in-
dicators forpeiiodonlal aitacbment loss in
a population of older community-dwelling
blacks and whites. Journal of Periodonio-
logy b\. 521 528.
Bergstrom, J. (1990) Oral hygiene compli-
ance and gingivitis expression in cigarette
smokers. Scandinavian J<nirnal of Denial
Researeh 98. 497-503.
Bergstrom, J. & Eliasson. S. (1987) Cigarette

smoking and alveolar bone hcighl in sub-
jects with a high standard of oral hygiene.
Journal of Clinical Periodontology. 14.
466-469.
Bergstrom. J.. Elia.sson, S. & Preber, H.
(1991) Cigarette smoking and periodontal
bone loss [published erratum appears in J
Periodontol 1991 Dec; 62(12): 809].
Journal of Periodoniology 62. 242-246.
Bergstrom, J. & Floderus-Myrhed. B. (1983)
Co-twin control study of the relationship
between smoking and some periodontal
disease factors. Comnmnity Dentistry and
Oral Epidemiology 11. 113-116.
Danielsen, B.. ManjI. F, Nagelkerke, N..
Fejerskov. O.. & Baelum, V (1990) Effect
of cigarelte smoking on the transition dy-
namics in experimental gingivitis. Journal
of Clinical Periodontology 17. 159-164.
Feldman. R. S.. Bravacos. J. S. & Rose, L.
(1983) Association between smoking and
different tobacco products and peri-
odonial disease indexes. Journal of Period-
ontology S. 4ii[^iil.
Goultschin. J.. Cohen, H. D., Donchin, M.,
Brayer. L. & Soskolne, W. A. (1990) As-
sociation of smoking with periodontal
treatment needs. Journal of PeriodoiUolo^y
61, 364-367.
Grossi, S. G.. Zambon. J. J.. Ho. A. W.
Koch. G.. Dunford. R. G., Machtei, E. E-.
Norderyd, O. M. & Genco, R. J. (1994)
Assessment of risk for periodontal disease.
I. Ri^k indicators for attachment loss.
Journal of Periodontology 65. 260-267.
Haber. J.. Wattles. J.. Crowley. M,. Mandell.
R,, Joshipura. K. & Kent. R. L. (1993)
Evidence for cigarette smoking as a major
risk factor tor periodontitis. Journal of
Periodontology 64. 16-23,
Hansen. B. F . Bjertness. E. & Gjermo, P.
(1990) Changes in periodontal disease in-
dicators in 35-year-old Oslo citizens from
1973 to 1984. Journal of Clinical Pcriodon-
tohgv 17. 249 254.
Herulf. G. (1951) On the margin;tl alveolar
ridge in students. A roentgenografic study.
Acta Genetiea et Staiistica Medica. II,
(Abstract).
Herulf. G. (1968) On the marginal alveolar
ridge in adults. Swedish Dental Journal f}\.
675 703.
Holm. G, (1994) Smoking as an additional
risk for tooth loss. Journal of Periodonio-
/»i;,i 65. 996-1001.
Homing. G. M.. Hatch. C. L., & Cohen, M.
E. (1992) Risk indicators for periodontitis
in a military treatment population. Journal
of Periodontology 63. 297-302.
Ismail. A. I.. Btirt, B. A. & Eklund. S. A.
(1983) Epidemiologic patterns of smoking
and periodontal disease in the United
Slates. Journal of the .American Dental As-
sociation 106. 617-623.
Linden, G. J, & Mullally. B. H. (1994) Ciga-
rette smoking and periodontal destruction
in young adults. Journal ol Fi'iicdomolo^y
65,'718-723.
Locker, D, Ford, J. & Leake. J. L. (1996)
Incidence of and risk factors for tuoih
loss in a population of older Canadians.
Journal of Dental Research 75, 783 789.
Locker. D. & Leake. J. L. (1993) Risk indi-
cators and risk markers for periodontal
disease experience in older adults living in-
dependently in Ontario. Canada. Journal
of Dental Research 11. 9 17.
MacFarlane. G. D., Herzberg. M. C . Wolff.
L. F & Hardie. N. A. (1992) Refractory
periodontitis associated with abnormal
polymorphoniiclear leukocyte phago-
cytosis and cigarette smoking. Journal of
Periodontology 63, 90K-913.
Machtei, F. E.. Dunford. R.. Hausmann.
E., Grossi. S. G.. Powell. J.. Cummins.
D., Zambon. J. J. & Genco. J. (1997)
Longitudinal study of prognostic factors
in established periodontitis patients.
Journal of Clinical PerioJontology 24,
102 109.
Martinez-Canut. P. Lorca. A. & Magan. R.
(1995) Smoking and periodonlal disease
severity. Journal of Clinical Periodontology
22. 734 749.
Massler. M. & Ludwick. W. (1952) Relation
of dental caries experience and gingivitis
to cigarette smoking in males 17 to 21
years old. Journal oj Denial Research 31.
319-322.
Mullally. B. H. & Linden. G. J. (1996) Molar
furcation involvement associated with
cigarette smoking in periodontal referrals.
Journal of Clinical Periodontology 23. 658
661.
Osterberg. T & Mellsirom. D. (1986) To-
bacco smoking; a major risk factor for los.s
of teeth in lliree 70-year-old cohorts. Com-
munity Dentistry ami Oral Epidemiology
14, 367-370.
Preber, H. & Bergstrom, J. (1985) Occurrence
of gingival bleeding in smoker and non-
smoker patients. Aeta Odontolo^ica Scan-
dinavica 43. 315-320.
Preber. H. & Berastrom. J, (1986) The effect
Smoking ami dental health status 305
of non-surgical treattnent on periodontal
pockets in smokers and non-smokers.
Journal of Clinical Periodontolof^y. 13.
319-323.
Preber, H. & Bergstrom, J. (1986) Cigarette
smoking in patients referred for peri-
odonta! treatment, Scandinavian Journal
of Dental Research 94. 102-108.
Preber. H.. Bergstrom. J. & Linder. L. E.
(1992) Occurrence of periopathogens in
smoker and non-smoker patients. Journal
of Clinical Periodoniolo^y. 19, 667 67!.
Ramljord. S, P (1959) Indices for prevalence
and incidence of periodontal disease.
Journal oJ Periodontolof^y. 30. 51-59.
Ravald. R . Birkhed. D. & Hamp. S. E.
(1993) Root caries susceptibility in peri-
odontally treated patients. Results after 12
years. Journal of Clinical Periodontology
20. 124-129,
Sakki. T, K,, Knuuttila. M. L.. Vimpari. S.
S.. & Hartikainen, M. S. (1995) Associ-
ation of lifestyle with periodontal health.
Community Dentistry and Oral Epidemi-
ology 23, 155-158.
Sheiham. A.. (1971) Periodontal disease and
oral cleanliness in tobaeco smokers.
Journal of Periodonrology 42. 259-263.
Soder. B., Jin. L. J.. Soder. P O. & Wikner.
S. (1995) Clinical characteristics of de-
structive periodontitis in a risk group of
Swedish urban aduhs. Swedish Dental
Journal 19, 9 15.
Stoltenberg, J. L.. Osborn. J. B.. Pihlstrom.
B. L.. Herzberg. M. C , Aeppli. D, M..
Wolff, L. F & Fischer. G, E. (1993) As-
sociation between cigarette smoking, bac-
terial pathogens, and periodontal status.
Journal of Periodontaiogy. 12. 1225-1230.
Sundberg. H. (1989) Dnr 5501 348,
Wiktorsson. A, M, (1995) Dental caries and
dental fluorosis among adults in two
Swedish communities with optimal and
low water fluoride concentrations. Thesis,
Faculty of Odontology. Karolinska Insti-
tute. Sweden,
Wolff, L,. Dahlen. G. & Aeppli. D, (1994)
Bacteria as risk markers for periodontitis.
Journal of Periodontology 65. 498 510,
Address;
Per A.\cls.son
Deparimeni of Preveniive Dentistry
Alvgatan 47
S-65: 30 Karlstad
Stye den