ArticlePDF Available

Abstract and Figures

To determine the cross-sectional association of the osteoporotic status of patients with the number of their teeth, with and without taking into account age and/or smoking. At four centres, the study recruited 665 females aged 45-70 years and the number of teeth was counted for 651 subjects. Bone density was measured at the total hip, femoral neck and lumbar spine. The mean number of teeth in the osteoporotic subjects was 3.3 fewer than normal subjects and 2.1 fewer if those with no teeth were excluded. The association between osteoporosis and having <6 or having <28 teeth remained significant after adjusting for age, smoking and centre with p-values of 0.016 and 0.011, respectively. A single regression model for tooth count with normal errors would not fit all the data. By fitting mixture regression models to subjects with tooth count >0, three clusters were identified corresponding to different degrees of tooth loss. The overall effect of osteoporosis was as follows: -1.8 teeth before and after adjusting for smoking, -1.2 teeth after adjusting for age, and -1.1 teeth after adjusting for both age and smoking. We have established a significant association between osteoporosis and tooth loss after adjusting the effect for age and smoking.
Content may be subject to copyright.
Tooth loss and osteoporosis: the
osteodent study
Nicopoulou-Karayianni K, Tzoutzoukos P, Mitsea A, Karayiannis A, Tsiklakis K,
Jacobs R, Lindh C, van der Stelt P, Allen P, Graham J, Horner K, Devlin H, Pavitt S,
Yuan J. Tooth loss and osteoporosis: the osteodent study. J Clin Periodontol 2009; 36:
190–197. doi: 10.1111/j.1600-051X.2008.01365.
Abstract
Aim: To determine the cross-sectional association of the osteoporotic status of
patients with the number of their teeth, with and without taking into account age
and/or smoking.
Material & Methods: At four centres, the study recruited 665 females aged 45–70
years and the number of teeth was counted for 651 subjects. Bone density was
measured at the total hip, femoral neck and lumbar spine.
Results: The mean number of teeth in the osteoporotic subjects was 3.3 fewer than
normal subjects and 2.1 fewer if those with no teeth were excluded. The association
between osteoporosis and having o6 or having o28 teeth remained significant after
adjusting for age, smoking and centre with p-values of 0.016 and 0.011, respectively.
A single regression model for tooth count with normal errors would not fit all the data.
By fitting mixture regression models to subjects with tooth count 40, three clusters
were identified corresponding to different degrees of tooth loss. The overall effect of
osteoporosis was as follows: !1.8 teeth before and after adjusting for smoking, !1.2
teeth after adjusting for age, and !1.1 teeth after adjusting for both age and smoking.
Conclusions: We have established a significant association between osteoporosis and
tooth loss after adjusting the effect for age and smoking.
Key words: bone mineral density;
osteoporosis; tooth loss
Accepted for publication 17 November 2008.
Osteoporosis is one of the commonest of
chronic diseases and is a disease in which
bone becomes porous and more suscepti-
ble to fracture. It is estimated that one in
three postmenopausal women and one in
five men over the age of 50 years are
affected (European Parliament Osteoporo-
sis Interest Group and EU Osteoporosis
Consultation Panel 2005). Osteoporosis is
adiseasethathasprovokedconsiderable
interest amongst dentists in the context of
its possible impact upon periodontal dis-
ease, residual alveolar ridge resorption
and implant success rates (Hildebolt
1997). In recent years, research has also
tried to determine whether dental radio-
graphic evidence of osteopenia may be
used as a way of identifying individuals at
risk of having osteoporosis (White 2005).
The aetiology of tooth loss is multi-
factorial, but one particular focus of
interest has been whether osteoporosis is
a contributory factor. Various researchers
have addressed this question, in studies of
varying qualities. The weight of the
evidence suggests that there is a signifi-
cant relationship between the number of
teeth present and skeletal bone mineral
density (BMD) (Krall et al. 1994, 1996,
Taguchi et al. 1995, 1999, Drozdzowska
et al. 2006), although two studies did
not demonstrate relationships (Earnshaw
et al. 1998, Taguchi et al. 2004). Simi-
larly, the majority of studies comparing
tooth number in osteoporotic and normal
subjects showed a smaller number in the
former group (Kribbs 1990, Mohammad
et al. 2003, Bodic et al. 2005, Yoshihara
et al. 2005), although conflicting results
have also been reported (May et al. 1995,
Mohammad et al. 1997).
From 2003 to 2005, the OSTEODENT
multicentre research project was carried
out with the aim of identifying the value
of a range of dental radiographic and
clinical indices for osteoporosis diagno-
sis in women. As part of this study,
subjects underwent a ‘‘gold standard’’
assessment of osteoporosis status using
dual energy X-ray absorptiometry
(DXA), collection of medical and life-
style information, and a panoramic den-
tal radiographic examination. This large
dataset has offered an opportunity to
conduct studies beyond the original focus
of the OSTEODENT project. The aim of
the study reported here was, therefore, to
determine whether the osteoporosis sta-
Kety Nicopoulou-Karayianni
1
,
Panagiotis Tzoutzoukos
1
, Anastasia
Mitsea
1
, Athanasios Karayiannis
1
,
Kostas Tsiklakis
1
, Reinhilde Jacobs
2
,
Christina Lindh
3
, Paul van der Stelt
4
,
Philip Allen
5
, Jim Graham
5
,
Keith Horner
6
, Hugh Devlin
6
,
Susan Pavitt
7
and Jingsong Yuan
8
1
Dental School, University of Athens, Athens,
Greece;
2
Oral Imaging Centre, School of
Dentistry, Oral Pathology and Maxillofacial
Surgery, Katholieke Universiteit Leuven,
Leuven, Belgium;
3
Faculty of Odontology,
Malmo University, Malmo
¨, Sweden;
4
Academic Centre for Dentistry, Amsterdam,
The Netherlands;
5
Division of Imaging
Science and Biomedical Engineering,
University of Manchester, Manchester, UK;
6
School of Dentistry, University of
Manchester, UK;
7
Clinical Trials Research
Unit, University of Leeds, UK;
8
School of
Mathematics, University of Manchester, UK
Conflict of interest and source of
funding statement
The authors declare that they have no
conflicts of interests.
This work was supported by a research and
technological development project grant
from the European Commission FP5
‘‘Quality of Life and Management of Liv-
ing Resources’’ (QLK6-2002-02243).
J Clin Periodontol 2009; 36: 190–197 doi: 10.1111/j.1600-051X.2008.01365.x
190 r2009 John Wiley & Sons A/S
tus of perimenopausal and postmenopau-
sal women was predictive of their num-
ber of teeth, independent of their age or
smoking status.
Material and Methods
Six hundred and sixty-five women, aged
45–70 years, were recruited into the
study in four European centres: Manche-
ster (UK), Leuven (Belgium), Malmo
(Sweden) and Athens (Greece). Local
ethical approval for the study was
obtained at each centre and all subjects
gave informed consent before inclusion
in the study. Of the 665 subjects, com-
plete data for the osteoporotic status, age
and number of teeth were known for 651
and the smoking status for 650 patients.
Subjects were recruited from the
patient and staff pool in the vicinity of
each institution, using local literature and
publicity available at each centre. The
group was mainly Caucasian, and insuffi-
cient numbers of other racial groups were
present for analysis. This methodology
tended to recruit a large proportion of
health-conscious individuals without
osteoporosis, so a stratified sampling pro-
cedure was used at one centre to recruit
osteoporotic individuals from a metabolic
disease clinic. This targeted recruitment
involved separate ethics approval to
approach patients with osteoporosis diag-
nosed within the last 6 months and, there-
fore, with minimal exposure to treatment
medication. The final percentage of osteo-
porotic individuals in the whole sample
was representative of other studies that
have examined a similar female age group
(WHO 1994).
Subjects were asked if they had ever
smoked cigarettes and were scored 0 for
never having smoked and 1 for having
smoking experience. Many factors affect
bone metabolism, but information about
age and smoking status of each patient
was collected because there is evidence
that these factors have a strong influence
on tooth retention.
The BMD of each subject was mea-
sured using DXA performed at three
different sites, lumbar spine (L1 to L4),
femoral neck and total hip, using DXA.
The precision of DXA measurements for
a variety of sites and devices has been
found to have a range of coefficients of
variation equal to 0.5–3% (Simpson &
Truscott 2000). These scans were per-
formed on either the Hologic QDR 4500,
Hologic Discovery (Hologic Inc., Bed-
ford, MA, USA) or the GE Lunar Prod-
igy (GE Lunar Corporation, Madison,
WI, USA). Standardized T-scores were
used (according to the WHO criteria for
Caucasian women) to classify women in
the sample as osteoporotic (T-score more
than 2.5 SD below the young adult mean
BMD) or normal/osteopenic (BMD
T-score 4!2.5SD). In the data analysis,
subjects were classified as osteoporotic if
they had osteoporosis at any measure-
ment site, and normal if otherwise.
Dental panoramic radiography was
performed on each subject using digital
and conventionally processed dental
panoramic radiography films. The Leuven
and Malmo centres used a Cranex DC3
machine (Soredex, Tuusula, Finland)
whereas the Athens and Manchester cen-
tres used a Planmeca PM2002CC machine
(Planmeca, Oy, Helsinki, Finland).
Tooth counting was done manually
using a simple MATLAB
s
program
(The MathWorks Ltd, Cambridge, UK)
which displayed each of the images in a
designated folder one at a time. For each
image, the user was prompted to enter
the numbers of incisor, canine, premolar
and molar teeth, and the results were
written to disc. The observer was blinded
as to the patient’s skeletal bone mineral
status. Roots, impacted teeth and implants
were excluded from the data analysis; for
inclusion, each tooth had to have at least
3mm of crown height. Occasionally,
buried roots were present but were
excluded, as in the study by Bollen et al.
(2004). There was no instance of a person
with a high caries rate and multiple roots.
Statistical analysis
The distribution of tooth count was
studied by examining plots against age
and histograms conditioning on osteo-
porotic status. A w
2
-test of independence
was conducted to assess the association
between osteoporosis and having a very
low or very high tooth count, which was
further analysed using logistic regres-
sion. Hypotheses of location shifts in the
distribution of tooth count due to osteo-
porosis were tested using Wilcoxon rank
sum tests and the amount of shift was
determined using non-parametric tests
of equal probability density functions.
The assumptions underlying standard
multiple regression model building were
not satisfied. Standard multiple regres-
sion models did not fit the data well,
with small R
2
(i.e., percentage of varia-
tion in the outcome variable explained
by the explanatory variables) and skewed
distribution of residuals resulting in
invalid conclusions. As a single regres-
sion model would not fit all the data,
mixtures of regression models (Turner
2000) were fitted to the dependent vari-
able, number of teeth, with subsets of
age, smoking status and osteoporosis
as explanatory variables. Mixture regres-
sion analysis may provide an insight into
hidden subgroups in data not recorded,
that further investigation may show as
reflecting characteristics such as lifestyle
and oral hygiene.
Results
The number of teeth was counted in 651
subjects, of whom 140 were osteoporotic.
AhistogramwasproducedinFig.1a
showing proportions of 0, 1, 2, up to 32
tooth counts in the osteoporotic group
(red) and in the normal group (blue). It
was observed that the proportion of low
tooth counts (o6) were higher and the
proportion of high tooth counts (427)
were lower in the osteoporotic group (no
tooth counts of 32), and that the distribu-
tion had excess 0s, a long tail towards 0
and was skewed to the left for both groups.
Testing the association between
osteoporosis and tooth count
A contingency table (Table 1) was con-
structed with osteoporosis status in rows
and whether a subject had fewer than
six teeth in columns. A w
2
-test of inde-
pendence was significant (w
2
516.78 on
1 df, po0.001), which showed an asso-
ciation between osteoporosis and having
fewer than six teeth.
A second contingency table (Table 2)
was constructed with osteoporosis status
in rows and whether a subject had 28 or
more teeth in columns. A w
2
-test of inde-
pendence was significant (w
2
514.24,
on 1 df, po0.001), which showed an
association between osteoporosis and
having fewer than 28 teeth. Increasing
the threshold to 29, 30 and 31 increased
the p-value to 0.011, 0.052 and 0.080,
respectively.
Using logistic regression, the associa-
tion between osteoporosis and having
few (o6) or not quite a full complement
of teeth (o28) remained significant after
adjusting for age, smoking and centre
with p-values of 0.016 and 0.011,
respectively.
Tooth loss and osteoporosis 191
r2009 John Wiley & Sons A/S
Testing location shift
Close examination of the histogram in
Fig. 1a showed that if the extreme left
tail of the distribution was ignored, the
distribution for the osteoporotic group
may be a shifted version of that for the
normal group. The sample medians of
23 and 25 suggested a location shift by
two teeth. Figure 1b shows what would
happen to the histogram if two extra
teeth were added to this group – it would
be a reasonable match between the two
distributions over the range 8–32. A
Wilcoxon rank sum test on a location
shift by two teeth was not significant
(p50.587) based on teeth count from 6
to 30 from the osteoporotic group and
those between 8 and 32 from the normal
group. For comparison, test results on
location shifts by 0, 1 and 3 teeth were
respectively significant (p50.001),
non-significant (p50.243) and border-
line significant (p50.051).
Identifying possible confounding
variables
The osteoporotic group apparently had,
on average, lower tooth counts than the
normal group, but the difference may
not be entirely attributable to osteoporo-
sis. The possibility of confounding by
smoking and age is considered below.
There were about the same number of
smokers as non-smokers among the
normal subjects (252 versus 258), and
among the osteoporotic subjects (75
versus 65). The mean number of teeth
among subjects who smoked was 22.34
(SE 50.394). Formally, this was not sig-
nificantly different (z-score of !1.94,
p50.053) from the mean tooth count
of 23.38 (SE 50.367) for subjects who
did not smoke. The median tooth count
was 25 for both smokers and non-smo-
kers. Excluding subjects with no teeth,
the conditional means were 23.19
(SE 50.323) and 23.83 (SE 50.327),
respectively, for smokers and non-smo-
kers, which were not significantly differ-
ent (p50.166). Smoking was, therefore,
not a confounding factor in the effect of
osteoporosis on tooth count which was
confirmed in regression analysis.
The mean age for the osteoporotic
group (59.24, SE 50.51) was signifi-
cantly higher (po0.001) than that of the
normal group (53.76, SE 50.25). Among
the younger subjects (age o57.5 years),
the effect of osteoporosis was two teeth as
measured by the median tooth count, and
among older subjects (age X57.5) the
difference in median tooth count was 1.
Thus, age may be a confounding variable
and the effect of age was adjusted for in
regression analysis.
Adjusting for age by mixture regression
The distribution of tooth count had a wide
range from 0 to 32 with a long tail
towards 0 and a left skew which remained
the case when conditioning on age or
osteoporosis status. Based on our assess-
ment of Fig. 2, we concluded that a single
regression model with normal errors
would not fit all the data, requiring a
mixture of regression models to be fitted.
Mixture regression is also known as
latent class regression (Wedel & DeSarbo
1995), where a latent variable represents
several unobserved classes and there is a
generalized linear regression model for
each class or cluster. Using the R package
Mixreg (Turner 2006), three clusters were
identified in the data (excluding the 0s) as
described by three linear regression mod-
els of tooth count on age and osteoporosis
status. Each subject originates from one of
three normal populations according to
probabilities p
1
,p
2
and p
3
which add up
to 1, and each distribution has a condi-
tional mean on age and osteoporosis in a
linear form. The parameters were esti-
mated by maximum likelihood via the
EM algorithm (Turner 2000). Each data
point was assigned to the class with the
largest conditional probability. The results
are shown in Fig. 3 with labels 1–3 in font
sizes proportional to the posterior prob-
abilities, and the fitted linear regression
line for each cluster is superimposed.
Details of the fitted regression models
are given in Table 3. The effect of
osteoporosis was not significant in clus-
ters 2 (p50.16) and 3 (p50.06), while
it was significant in cluster 1 (p50.03).
For cluster 1 which corresponded to
high tooth counts (mean 527) and cov-
ered 55.7% of the study population,
osteoporosis accounted for nearly one
(0.85) less tooth on average after adjust-
ing for age. Increasing age was asso-
ciated with reduced numbers of teeth
except for cluster 3 (p50.64) where the
difference in mean tooth count due to
osteoporosis was 9.5 !752.5.
The regression lines may not appear
the best fit for each cluster but the
cluster memberships are estimated
and thus not definitive. The same
equations can be obtained by weighted
least squares regression using posterior
probabilities as weights. For each
data point there were three fitted values,
one from each regression equation,
and thus three residual values. The
residuals were standardized for diagnos-
tic purposes. Figure 4 shows fitted
values against observed values of tooth
counts, standardized residuals against
age and against fitted values, and a
normal quantile plot for the standardized
residuals. Ignoring the small dots, which
represent small posterior probabilities,
and concentrating on the larger symbols
(Turner 2000) we found no over-
whelming evidence against model
adequacy.
Fig. 1. (a ) Histogram of tooth count for osteoporotic group (red) and that for normal group
(blue). (b) Histogram of tooth count +2 for the osteoporotic group and tooth count for normal
group.
Table 1. Cross tabulation of osteoporotic status
with number of teeth (o6 or X6 teeth)
o6 Teeth X6 Teeth Total
Osteoporotic 14 126 140
Normal 12 499 511
Total 26 625 651
Table 2. Cross tabulation of osteoporotic status
with number of teeth (o28 or X28 teeth)
X28 Teeth o28 Teeth Total
Osteoporotic 19 121 140
Normal 150 361 511
Total 169 482 651
192 Nicopoulou-Karayianni et al.
r2009 John Wiley & Sons A/S
Using posterior probabilities as wei-
ghts, the three fitted values were com-
bined into one, which is the estimated
conditional mean. These are plotted in
Fig. 5 against observed values. The
mixture regression models explained the
data very well with a pseudo-R
2
value of
88.5%. However, they have limited pre-
dictive power for cluster memberships/
tooth counts.
The additive mixture regression mod-
el with only main effects (no interac-
tion) was satisfactory, suggesting that
effect modification was not important.
Adding interactions between age and
osteoporosis allowed different gradients
within each cluster but resulted in a non-
significant increase (p50.33, df 53) in
log-likelihood value from !1875.024
to !1873.314. Adding main effect
terms for smoking (and excluding one
subject with missing smoking status
data) did not result in a significant
increase (p50.35, df 53) in log-like-
lihood value (from !1872.759 to
!1871.131) either.
We combined data from all four
centres in order to have a sufficiently
large dataset. Although the model
(Table 3) did not include the recruitment
centre as a variable, it fitted data from
each centre very well (Fig. 6).
The effects of osteoporosis in each
cluster were summarized in Table 3,
from which an overall effect can be
calculated by averaging the effects for
each cluster using cluster probabilities
as weights. The overall effect of osteo-
porosis was found to be !1.8 teeth
before and after adjusting for smoking,
!1.2 teeth after adjusting for age, and
!1.1 teeth after adjusting for both age
and smoking. These results agreed well
with the location shift of two teeth found
earlier and confirmed the necessity to
adjust the effect of osteoporosis by age,
but not by smoking.
The mixture of regressions with nor-
mal errors are at best a good approxima-
tion as strictly speaking tooth count is a
discrete random variable and there was
some evidence of non-linear depen-
dence on age. Alternative models with
binomial or Poisson errors (parametric
or semi-parametric, with or without in-
corporating zero inflation) can be fitted
under the somewhat unrealistic assump-
tion of independent loss at constant rate
and with knowledge of the amount and
duration of tooth loss, but the normal
mixture regression was by far the best
interpretable model.
Discussion
Osteoporosis and periodontitis are both
chronic diseases presenting several
similarities. Both involve bone resorp-
tion, with common risk factors such as
age, smoking, systemic disease and diet-
ary factors. Calcium absorbed from
the diet is difficult to measure in large
populations, but supplementary calcium
and vitamin D, aimed at preventing
osteoporosis, has also been shown to
reduce tooth loss (Krall et al. 2001).
Oestrogen therapy also protects both
Fig. 3. Mixture regression model with the fitted linear regression line for each of the three
clusters superimposed. Each data point was assigned to the class with the largest conditional
probability with labels 1 to 3 in font sizes proportional to the posterior probabilities.
Fig. 2. Plots of tooth count against age for osteoporotic (red) and normal subjects (blue) with
least squares (lower) and robust (higher) regression lines superimposed. The robust estimator
(Huber 1981) calculated using the rim package (Venables and Ripley, 2002) was less affected
by data points with low tooth counts.
Table 3. Regression coefficients (Est) for independent variables in each of the three clusters with
associated standard errors (SE)
Cluster 1 (p
1
50.557) Cluster 2 (p
2
50.367) Cluster 3 (p
3
50.076)
Const Age Op(1) Const Age Op(2) Const Age Op(3)
Est. 32.07 !0.09 !0.85 33.59 !0.22 !1.37 13.33 !0.06 !2.84
SE 1.46 0.03 0.38 3.54 0.07 0.98 7.06 0.12 1.50
p0.00 0.00 0.03 0.00 0.00 0.16 0.06 0.64 0.06
Op (1), Op (2) and Op (3) represent regression coefficients of osteoporosis (Op) for each cluster
using the predictor variables in the table. Each subject originates from one of three normal
populations according to probabilities p
1
,p
2
and p
3
, which add up to 1.
Const: constant, p: level of probability. The bold numerals emphasize the important numbers in this
table. These important numbers refer to osteoporosis.
Tooth loss and osteoporosis 193
r2009 John Wiley & Sons A/S
lumbar spine and mandibular bone den-
sity (Jacobs et al. 1996).
We have shown in this study that
osteoporosis at either the hip or spine
is a risk factor for tooth loss. The overall
effect of osteoporosis was one or two
teeth lost depending on whether age was
taken into account. Severe osteoporosis
may be a co-factor in alveolar bone loss
(Phillips & Ashley 1973, Ward & Man-
son 1973). Study populations that have
used young, perimenopausal women
(aged 46–55 years), where osteoporosis
is less prevalent, are more likely to fail
to find any association (Elders et al.
1992). Another study of an older popu-
lation of women (aged 65–76 years) was
unable to find a significant association
between self-reported tooth loss and
BMD at the hip and spine, but there
may have been errors in the accuracy of
self-reporting (May et al. 1995).
Some studies have reported an asso-
ciation between osteoporosis and tooth
loss, for example, a significant associa-
tion between risk of hip fracture and
tooth counts (Astrom et al. 1990) or that
the lumbar spine BMD was significantly
lower among patients who acquired
dentures at age 40 or earlier (Krall
et al. 1994). Osteoporosis may influence
the rate of bone loss in chronic perio-
dontitis (Taguchi et al. 1995) which may
explain the greater percentage of eden-
tulous subjects found in some studies
(Kribbs 1990, Mohammad et al. 2003).
Yoshihara et al. (2004) found a weak
relationship between BMD and perio-
dontal disease progression, although
it was statistically significant, and in a
later study found a correlation between
BMD of the os calcis and the number of
remaining teeth (Yoshihara et al. 2005).
Osteoporotic women have a higher risk
of tooth loss, and may undergo greater
bone resorption after tooth loss, com-
pared with healthy women of the same
age range (Bodic et al. 2005).
In our large study population, a
statistically significant relationship was
Fig. 4. Fitted values are plotted against observed values of tooth counts, standardized residuals against age and against fitted values, and a
normal quantile plot shown for the standardized residuals. The size of the dots represents the posterior probability, and the color denotes the
cluster or equation used to calculate the fitted value.
Fig. 5. Combined fitted values against observed data with dotted lines at "5 teeth.
194 Nicopoulou-Karayianni et al.
r2009 John Wiley & Sons A/S
found between the total number of teeth
and osteoporotic status. It must be
emphasized that the effect of osteoporo-
sis may be of minor importance in
determining tooth loss, with other
clinical and socioeconomic factors play-
ing a more influential role. These factors
affect the prevalence of periodontal
disease; therefore, they may have influ-
enced our tooth loss results. The impor-
tant roles of patient age and periodontal
status have been previously demon-
strated (Mohammad et al. 1997) because
when these factors were included, tooth
loss was not significantly different be-
tween subjects with low and high spinal
bone density. They concluded that total
tooth loss was not directly associated
with systemic bone density. Age and
smoking have been shown in regression
analysis to have a strong influence on
tooth number in an elderly population,
whereas a history of osteoporotic frac-
ture was not significant (Bollen et al.
2004). Regular maintenance treatment
in a cross-section of highly motiv-
ated subjects with chronic periodontitis
seemed to be successful in preventing
progressive periodontal tissue destruc-
tion in current smokers (Fisher et al.
2008). The sample in our study was a
highly motivated group. In our study,
one alternative explanation of the lack
of effect of smoking on tooth loss may
be due to the selection of subjects and
how the smoking variable was defined.
In our analysis, we did not distinguish
whether subjects were current smokers
or had ceased to smoke decades
previously. This may have given rise
to a misclassification bias. A previous
publication (Karayianni et al. 2007)
described the BMD and clinical data
from our sample in more detail. The
Osteoporosis Index of Risk Assessment
(OSIRIS) was used which combines
information about a patient’s age,
weight, their hormone replacement ther-
apy and any history of low trauma
fracture as clinical risk factors in detect-
ing osteoporosis.
Low serum oestrogen has been shown
to increase skeletal (Devlin et al. 1990)
and alveolar bone resorption (Binte
Anwar et al. 2007) in ovariectomi-
zed animals, with destruction of the
trabeculae. Further research is required
to distinguish this phenomenon from
plaque-induced periodontal disease;
some clinical human studies have shown
no significant association between
periodontal disease and systemic BMD
(Weyant et al. 1999, Famili et al. 2005).
The aetiology of the osteoporosis-
induced atrophy of alveolar bone is poorly
understood, but fewer teeth have been
found in those with low systemic skeletal
bone density (Inagaki et al. 2005).
In the present study, whether the
patients smoked, their age and osteoporo-
tic status were chosen as explanatory
variables because these can be measured
reliably and accurately. Inclusion of addi-
tional factors in the regression analysis
such as body mass index, calcium intake
and menopausal status that might affect
tooth number is complicated by the
collinearity that this inevitably introduces.
Collinearity occurs when explanatory
variables are significantly correlated with
one another, causing an unreliable model
that is difficult to interpret.
There are quite possibly various impor-
tant differences across the nationalities/
cultures in the four countries that may
affect either/both tooth loss and osteo-
porosis. Such factors may include differ-
ences in nutritional intakes of caffeine,
calcium-rich foods and alcohol, all factors
found to be associated with osteoporosis
risk as well as tooth loss risk. In addition,
there may be important geographic varia-
tions in solar exposure patterns, which is
the principal source of variation in vita-
min D levels in many cultures. As can be
seen from Fig. 6, the number of osteo-
porotic cases was small for centres 2, 3
and 4, and the numbers belonging to
cluster 3 are 0, 1 and 3 in these centres,
respectively. This means that models
with an additional variable for centre
could not be reliably fitted to the data.
Fig. 6. The model was based on data from all four centres and although it did not include the recruitment centre as a variable it fitted data from
each individual centre very well.
Tooth loss and osteoporosis 195
r2009 John Wiley & Sons A/S
We could not improve the fit by adjusting
the regression lines using a centre vari-
able because there are so few data points
around some of the regression lines.
In conclusion, we have shown that
after adjusting for smoking status and
age, those with osteoporosis had fewer
teeth than subjects with a normal/osteo-
penic BMD. Further separate longitudi-
nal studies are planned to determine a
causative basis for this finding, which is
not possible from data in the present
cross-sectional study.
References
Astrom, J., Backstrom, C. & Thidevall, G.
(1990) Tooth loss and hip fractures in the
elderly. Journal of Bone and Joint Surgery
72-B, 324–325.
Binte Anwar, R., Tanaka, M., Kohno, S.,
Ikegame, M., Watanabe, N., Nowazesh Ali,
M. & Ejiri, S. (2007) Relationship between
porotic changes in alveolar bone and spinal
osteoporosis. Journal of Dental Research 86,
52–57.
Bodic, F., Hamel, L., Lerouxel, E., Basle, M. F.
& Chappard, D. (2005) Bone loss and teeth.
Joint, Bone, Spine 72, 215–221.
Bollen, A. M., Taguchi, A., Hujoel, P. P. &
Hollender, L. G. (2004) Number of teeth and
residual alveolar ridge height in subjects with a
history of self-reported osteoporotic fractures.
Osteoporosis International 15, 970–974.
Devlin, H., Ferguson, M. W. J. & Carter, H. D.
(1990) Cancellous bone resorption in the
proximal ilium of the ovariectomized rat.
Calcified Tissue International 46, 395–400.
Drozdzowska, B., Pluskiewicz, W. & Michno,
M. (2006) Tooth count in elderly women in
relation to their skeletal status. Maturitas 55,
126–131.
Earnshaw, S. A., Keating, N., Hosking, D. J.,
Chilvers, C. E., Ravn, P., Mc Clung, M. &
Wasnich, R. D. (1998) Tooth counts do not
predict bone mineral density in early post-
menopausal Caucasian women. EPIC study
group. International Journal of Epidemiology
27, 479–483.
Elders, P. J., Habets, L. L., Netelenbos, J. C.,
van der Linden, L. W. & van der Stelt, P. F.
(1992) The relation between periodontitis and
systemic bone mass in women between 46
and 55 years of age. Journal of Clinical
Periodontology 19, 492–496.
European Parliament Osteoporosis Interest
Group and EU Osteoporosis Consultation
Panel (2005) Osteoporosis in Europe:
Indicators of Progress. Nyon, Switzerland:
International Osteoporosis Foundation, p. 2.
Famili, P., Cauley, J., Suzuki, J. B. & Weyant,
R. (2005) Longitudinal study of periodontal
disease and edentulism with rates of bone
loss in older women. Journal of Perio-
dontology 76, 11–15.
Fisher, S., Kells, L., Picard, JP., Gelskey, SC.,
Singer, DL., Lix, L. & Scott, DA. (2008)
Progression of periodontal disease in a main-
tenance population of smokers and non-
smokers: a 3-year longitudinal study. Journal
of Periodontology 79, 461–468.
Hildebolt, C. F. (1997) Osteoporosis and oral
bone loss. Dentomaxillofacial Radiology 26,
3–15.
Huber, P. J. (1981) Robust Statistics, pp. 1–19.
New York: Wiley.
Inagaki, K., Kurosu, Y., Yoshinari, N., Noguchi,
T., Krall, E. A. & Garcia, R. I. (2005) Efficacy
of periodontal disease and tooth loss to screen
for low bone mineral density in Japanese
women. Calcified Tissue International 77,
9–14.
Jacobs, R., Ghyselen, J., Koninckx, P. & van
Steenberghe, D. (1996) Long-term bone mass
evaluation of mandible and lumbar spine in a
group of women receiving hormone replace-
ment therapy. European Journal of Oral
Science 104, 10–16.
Karayianni, K., Horner, K., Mitsea, A., Bour-
kas, L., Mastouris, M., Jacobs, R., Lindh, C.,
van der Stelt, P. F., Harrison, E., Adams, J. E.,
Pavitt, S. & Devlin, H. (2007) Accuracy in
osteoporosis diagnosis of a combination of
mandibular cortical width measurement on
dental panoramic radiographs and a clinical
risk index (OSIRIS): the OSTEODENT
project. Bone 40, 223–229.
Krall, E. A., Dawson-Hughes, B., Papas, A. &
Garcia, R. I. (1994) Tooth loss and skeletal bone
density in healthy postmenopausal women.
Osteoporosis International 4, 104–109.
Krall, E. A., Gracia, R. I. & Dawson-Hughes, B.
(1996) Increased risk of tooth loss is re-
lated to bone loss at the whole body, hip,
and spine. Calcified Tissue International 59,
433–437.
Krall, E. A., Wehler, C., Garcia, R. I., Harris, S. S.
& Dawson-Hughes, B. (2001) Calcium and
vitamin D supplements reduce tooth loss in
the elderly. The American Journal of Medi-
cine 111, 452–456.
Kribbs, P. J. (1990) Comparison of mandi-
bular bone in normal and osteoporotic
women. Journal of Prosthetic Dentistry 63,
218–222.
May, H., Reader, R., Murphy, S. & Khaw, K. T.
(1995) Self-reported tooth loss and bone
mineral density in older men and women.
Age and Ageing 24, 217–221.
Mohammad, A. R., Bauer, R. L. & Yeh, C. K.
(1997) Spinal bone density and tooth loss in
a cohort of postmenopausal women. Inter-
national Journal of Prosthodontics 10,
381–385.
Mohammad, A. R., Hooper, D. A., Vermilyea,
S. G., Mariotti, A. & Preshaw, P. M. (2003)
An investigation of the relationship between
systemic bone density and clinical periodontal
status in post- menopausal Asian-American
women. International Dental Journal 53,
121–125.
Phillips, H. B. & Ashley, F. P. (1973) The
relation between periodontal disease and a
metacarpal bone index. British Dental Journal
134, 237–239.
Simpson, D. S. & Truscott, J. G. (2000) Meth-
odological and reporting considerations In:
Fordhom, J. N. (ed). Manual of Bone Densi-
tometry Measurements, p. 40. London:
Springer-Verlag.
Taguchi, A., Fujiwara, S., Masunari, N. &
Suzuki, G. (2004) Self-reported number of
remaining teeth is associated with bone
mineral density of the femoral neck, but not
of the spine, in Japanese men and women.
Osteoporosis International 15, 842–846.
Taguchi, A., Suei, Y., Ohtsuka, M., Otani, K.,
Tanimoto, K. & Hollender, L. G. (1999)
Relationship between bone mineral density
and tooth loss in elderly Japanese women.
Dentomaxillofacial Radiology 28, 219–223.
Taguchi, A., Tanimoto, K., Suei, Y., Otani, K.
& Wada, T. (1995) Oral signs as indicators
of possible osteoporosis in elderly women.
Oral Surgery, Oral Medicine, Oral Pathol-
ogy, Oral Radiology and Endodontics 80,
612–616.
Turner, T. R. (2000) Estimating the rate of
spread of a viral infection of potato plants
via mixtures of regressions. Applied Statistics
49 (Part 3), 371–384.
Turner, T. R. (2006) Mixreg: Functions to fit
mixtures of regressions. R package version
0.0-2. http://www.math.unb.ca/ #rolf/
Venables, W. N. & Ripley, B. D. (2002) Mod-
ern Applied Statistics with S, 4th edition.
Heidelberg: Springer.
Ward, V. J. & Manson, J. D. (1973) Alveolar
bone loss in periodontal disease and the
metacarpal index. Journal of Periodontology
44, 763–769.
Wedel, M. & DeSarbo, W. S. (1995) A mixture
likelihood approach for generalised linear
models. Journal of Classification 12, 21–55.
Weyant, R. J., Pearlstein, M. E., Churak, A. P.,
Forrest, K., Famili, P. & Cauley, J. A. (1999)
The association between osteopenia and
periodontal attachment loss in older women.
Journal of Periodontology 70, 982–991.
White, S. C. (2005) Oral radiographic predictors
of osteoporosis. Dentomaxillofacial Radiology
31, 84–92.
World Health Organization (1994) Assessment
of fracture risk and its application to screening
for postmenopausal osteoporosis. Technical
Report Series 843, WHO, Geneva.
Yoshihara, A., Seida, Y., Hanada, N., Kanashima,
K. & Myiazaki, H. (2005) The relationship
between bone mineral density and the number
of remaining teeth in community-dwelling
older adults. Journal of Oral Rehabilitation
32, 735–740.
Yoshihara, A., Seida, Y., Hanada, N. &
Myiazaki, H. (2004) A longitudinal study of
the relationship between periodontal disease
and bone mineral density in community-
dwelling older adults. Journal of Clinical
Periodontology 31, 680–684.
Address:
Prof. Hugh Devlin
School of Dentistry
University of Manchester
Higher Cambridge ST. Manchester
M15 6HF
UK
E-mail: hugh.devlin@manchester.as.uk
196 Nicopoulou-Karayianni et al.
r2009 John Wiley & Sons A/S
Clinical Relevance
Scientific rationale for the study:
Previous research has shown an
inconsistent effect of osteoporosis
on tooth loss. We used a large Eur-
opean dataset of 665 subjects, taking
into account a confounding factor
such as age. Statistical analysis used
mixture regression models.
Principal findings: The overall effect
of osteoporosis was found to be a
reduction of about 1.8 teeth before
and after adjusting for smoking, and
1.2 fewer teeth after adjusting for
age. The effect of smoking on tooth
loss was not significant in addition to
osteoporosis status.
Practical implications: Osteoporosis
was associated with a significantly
fewer number of teeth in this sample
of females.
Tooth loss and osteoporosis 197
r2009 John Wiley & Sons A/S
... The exact and practical definition of Osteoporosis by the World Health organization (WHO) is: T-score < -2.5 standard deviation (SD). T-score > -1 SD expresses normal BMD and -1 SD > T-score > -2.5 SD indicates osteopenia [1], [2], [3], [4] . The maxillary and mandibular bones are not deprived of osteoporosis and suffer from reduced bone mass and poor bone quality [3] . ...
... It's noticeable that not all researches revealed a positive correlation between osteoporosis and periodontitis [32] . Several lectures proved that there is a positive relationship between jaw bones and decreases of systemic skeletal BMD in osteoporosis [4] . Studies determined a positive correlation between mandibular BMD and lumbar spine, forearm, and femoral neck as the main sites in osteoporosis [3] . ...
... May et al. had a study that demonstrated a relationship between tooth loss and BMD in men, but it was a questionnaire study [34] . In an experimental study, after managing of smoking and age conditions, there was a relationship between tooth number and osteoporosis condition [4] . Darcey et al. study indicated that positive osteoporosis, smoking, and also rising age demonstrated relation with decreasing tooth numbers [3] . ...
Article
Full-text available
Osteoporosis can both, directly and indirectly, result in tooth loss.
... Clinical literature on osteoporosis has also documented a significant relationship between age-related bone loss and AMTL in living populations (Nicopoulou-Karayianni et al., 2009;Savi c Pavičin et al., 2017). Postcranial bone loss in patients diagnosed with osteoporosis and osteopenia is associated with alveolar bone loss in the mandible and maxilla (Guiglia et al., 2013;Jonasson & Rythén, 2016;Wactawski-Wende, 2001;Wang & McCauley, 2016). ...
... This loss in alveolar bone leads to a reduction in structural support for the tooth and can lead to premature exfoliation of teeth in these individuals. As a result, women diagnosed with osteoporosis are significantly more likely to experience tooth loss than non-osteoporotic women (Nicopoulou-Karayianni et al., 2009;Savi c Pavičin et al., 2017). ...
... Moreover, this finding is in line with the results of Klemetti et al. (1993aKlemetti et al. ( , 1993b, who found mandibular cortical BMD correlated well with postcranial cortical BMD, while mandibular trabecular BMD showed no significant relationship with postcranial BMD. The moderate effect of cortical BMD specifically on AMTL in the Point of Pines Pueblo sample is also consistent with clinical studies demonstrating a significant relationship between mandibular bone loss, periodontal bone loss, osteoporosis or osteopenia and tooth loss (Guiglia et al., 2013;Jang et al., 2015;Makker et al., 2012;Nicopoulou-Karayianni et al., 2009;Passos et al., 2013;Sultan & Rao, 2011). Future studies specifically including mandibular BMD, as well as periodontal disease and alveolar resorption, would help clarify this relationship. ...
Article
Previous archaeological research on dental health in the New World has documented significant sex differences in antemortem tooth loss (AMTL), with a much higher rate of AMTL in females versus males, particularly during the transition to agriculture. While AMTL can be caused by multiple factors, including periodontal disease, attrition, trauma, and cultural influences, sex differences are often attributed to the impact of female reproductive biology on oral health. Clinical research on osteoporosis has documented a significant relationship between AMTL and age‐related bone loss, which disproportionately affects women. However, this relationship has not been systematically investigated in prehistoric populations. This study aims to address this issue by investigating the relationship between sex, AMTL, and age‐related bone loss in an archaeological sample from East‐Central Arizona. AMTL, dental caries, and radial and femoral cortical and trabecular bone mineral density (BMD) were measured in individuals from Point of Pines Pueblo, Arizona (AD 1200‐1450). Our results revealed that while there was no statistically significant difference in AMTL between males and females in this sample, there were notable sex differences in the relationship between AMTL, caries, age, and BMD. There was a significant association between caries, age, and AMTL in females, but not in males. Conversely, while age had a significant effect on caries in males, there was no corresponding relationship in females. Cortical BMD had a moderate effect on AMTL in females, comparable to the effect of age, although this did not reach statistical significance. There was no significant effect of BMD on AMTL in males. The results suggest that biocultural processes differentially affected oral health in males and females at Point of Pines Pueblo, and that age‐related cortical bone loss potentially impacted AMTL in females in this population, but further research is needed.
... Οι Nicopoulou-Karayanni et al. (2009) αναφέρουν ότι η οστεοπόρωση στο ισχίο και της ΟΜΣΣ είναι επιδραστικός παράγοντας για την απώλεια των δοντιών, σε σύνολο ενός με δύο δοντιών σε σχέση με την ηλικία του ασθενούς [6] . Οι Yoshihara et al. (2004), έδειξαν μικρή σχέση μεταξύ της BMD και της υποτροπής των περιοδοντικών νόσων, και σε μια πιο πρόσφατη έρευνά τους, αναφέρουν σχέση μεταξύ BMD του οστού της πτέρνας και των οδόντων που έχουν απομείνει [7] Βέβαια, διαφέρει η απώλεια οδόντων μεταξύ των διαφόρων δουλειών, σε σχέση με τη δίαιτα και τη λήψη ασβεστίου, την πρόσληψη καφεΐνης, την κατανάλωση αλκοόλ, αίτια στενά συνδεδεμένα με την οστεοπόρωση. ...
... Οι Yoshihara et al. (2004), έδειξαν μικρή σχέση μεταξύ της BMD και της υποτροπής των περιοδοντικών νόσων, και σε μια πιο πρόσφατη έρευνά τους, αναφέρουν σχέση μεταξύ BMD του οστού της πτέρνας και των οδόντων που έχουν απομείνει [7] Βέβαια, διαφέρει η απώλεια οδόντων μεταξύ των διαφόρων δουλειών, σε σχέση με τη δίαιτα και τη λήψη ασβεστίου, την πρόσληψη καφεΐνης, την κατανάλωση αλκοόλ, αίτια στενά συνδεδεμένα με την οστεοπόρωση. Τέλος, σημασία έχει η έκθεση στον ήλιο ως κύριου παράγοντα μετατροπής ης προβιταμίνης D σε ενεργό βιταμίνη [4,[6][7][8][9] . ...
Article
Full-text available
Η παρούσα ανασκόπηση στοχεύει στην αξιολόγηση της ακρίβειας των διαφόρων πανοραμικών μορφομετρικών δεικτών για τη διάγνωση της οστεοπενίας/οστεοπόρωσης, με βάση μετα-αναλύσεων της ευαισθησίας και της ειδικότητάς τους και όπως αυτή μπορεί να συγκριθεί με πιο συστηματικές μεθοδολογίες, όπως η τεχνική της απορροφησιομετρίας διπλοενεργειακής δέσμης ακτίνων-X, DXA (Dual-energy X-ray Absorptiometry). Πραγματοποιήθηκε μια ηλεκτρονική αναζήτηση στα PubMed, Medline και συμπεριλήφθηκαν κλινικές δοκιμές, καθώς και ανασκοπήσεις. Αναφέρονται επίσης, κλινικές εκδηλώσεις και ακτινογραφικά ευρήματα, με σκοπό την καλύτερη κατανόηση. Η πανοραμική ακτινογραφία, ορθοπαντομογράφημα, είναι μια ακτινογραφία, που επιτρέπει την ακτινογραφική απεικόνιση σε δισδιάστατο επίπεδο, των οδόντων, των γνάθων του προσώπου, των ιγμορείων άντρων και λοιπών ανατομικών στοιχείων από το τράχηλο και πάνω και από τη ρινική κοιλότητα και κάτω. Η οστεοπενία είναι μια, αφορά μία μη φυσιολογική κατάσταση, κατά την οποία η οστική πυκνότητα είναι μικρότερη από το υγιές και χρήζει περαιτέρω διερεύνησης, σε κλινικό και ακτινογραφικό επίπεδο. Η μη αντιμετώπισή σε έγκαιρο χρονικό διάστημα, μεταπίπτει σε μια πιο χρόνια και επιβλαβή για τον οργανισμό κατάσταση, την οστεοπόρωση. Στη συνέχεια της ανασκόπησης, θα αναφερθούν οι τρόποι με τους οποίους μπορεί να γίνει η διάγνωση και η αντιμετώπιση αυτών των περιπτώσεων και πως η πανοραμική μπορεί να χρησιμοποιηθεί ως εργαλείο διάγνωσης.
... OP in the mandible is a bone disease characterized by the mass bone loss and fragility [41]. OP impairs not only the tooth [42] and periodontal tissues [43] but also implantation treatment, especially, causing marginal bone loss around implants [44]. Recently, mounting reports have proved that miRNAs have an effect on osteogenic differentiation of mesenchymal stem cells through transcriptional activation or inhibition of osteogenesis-related genes [45] [46] or the genes in some signal pathway [19]. ...
Article
Full-text available
Background Osteoporosis affects the mandible resulting in bone loss. Though impairments are not life threatening, they affect a person's quality-of-life particularly vulnerable elderly. MicroRNAs (miRNAs) are novel regulatory factors that play an important role in regulating bone metabolism. Autophagy is evolutionarily conserved intracellular self-degradation process and is vital in the maintenance of both miRNA and bone homeostasis. However, the role of autophagy in the pathogenesis of miRNA regulating osteoporosis remains unclear. Methods In the study, we established a rat osteoporosis model induced by ovariectomy (OVX) and isolated mesenchymal stem cells from mandible (MMSCs-M). Several miRNAs were identified to regulate osteoporosis in some studies. qRT-PCR was applied to examine the expression of miRNA, autophagy and osteogenic differentiation-related genes. Western blotting assays were performed to detect the expression of autophagy and osteogenic differentiation proteins. Immunofluorescence and transmission electron microscope were used to verify the autophagy activity. Transfecting technology was used to enhance or suppress the expression of miR-152-5p which enable us to observe the relationship between miR-152-5p, autophagy and osteogenic differentiation. Additionally, the measurement of reactive oxygen species was used to investigate the mechanism of autophagy affecting osteogenic differentiation. Results We found an upregulated expression of miR-152-5p in MMSCs-M in OVX group. Downregulated autophagy-related gene, proteins and autophagosome were detected in vitro of OVX group compared with sham group. Moreover, downregulation of miR-152-5p promoted osteogenic differentiation of MMSCs-M as well as enhanced autophagy-related proteins in OVX group. Conversely, overexpression of miR-152-5p showed opposite effect in sham group. Meanwhile, we found Atg14 (autophagy-related protein homolog 14) was identified to be a direct target of miR-152-5p theoretically and functionally. In other words, we confirmed inhibition of miR-152-5p promoted the osteogenic differentiation via promoting ATG14-mediated autophagy. Furthermore, miR-152-5p/ATG14-mediated autophagy regulated osteogenic differentiation by reducing the endogenous ROS accumulation and maintaining cellular redox homeostasis. Conclusion Our data suggest that miR-152-5p is the first identified to regulate osteogenic differentiation by directly targeting autophagy-related protein ATG14 and regulating oxidative stress and therapeutic inhibition of miR-152-5p may be an efficient anabolic strategy for osteoporosis.
... PMI is not the only factor taken into account in the craniofacial region; there are others. So far, the most extensive study using several existing methods of mandibular bone density classification is the multi-center project OSTEODENT [37][38][39][40][41][42]. The research carried out as part of the project did not indicate a method of assessing a panoramic X-ray that could effectively replace DXA in the diagnosis of osteopenia and osteoporosis; however, it highlighted the potential of dental radiology in identifying patients with low bone mass and the need for further research in this area. ...
Article
Full-text available
Background: The study aimed to evaluate radiomorphometric indices derived from panoramic X-rays and selected blood markers of bone turnover and neutrophil extracellular traps, with a view to identifying hemophilic patients at risk of developing osteoporosis. Methods: The study consisted of 50 adult men with hemophilia A and B (mild, moderate, and severe). The control group consisted of 25 healthy adult men. In both groups, blood samples were collected to determine concentrations of citrullinated histone H3 (CH3) and osteocalcin (BGLAP) with ELISA tests, and panoramic X-rays were obtained. Images were imported into AudaXCeph software to calculate two radiomorphometric indices: mental index (MI) and panoramic mandibular index (PMI). Concentrations of BGLAP and CH3 were compared with MI and PMI values in patients with and without hemophilia. Results: There were statistically significant differences in BGLAP, CH3, and PMI between the study and the control group (p < 0.05). Multivariate logistic regression analysis showed a predictive value for PMI, BGLAP, and CH3.The ROC curve with cutoff point (Youden index) at 0.40-PMI was calculated. No correlation was observed for the PMI index in any particular subgroup of patients. No correlation between MI and BGLAP/CH3 was observed. Conclusions: Simultaneous use of PMI value and BGLAP and CH3 levels may allow the identification of patients with hemophilia who requirea detailed diagnosis of osteoporosis with DXA.
... Το 2009 οι Karayianni et al. με βάση τα υπάρχοντα δεδομένα εξέτασαν την πιθανή συσχέτιση μεταξύ αριθμού δοντιών στο στόμα και διαγνωσμένης οστεοπόρωσης [13] . Συνυπολο- [14][15][16][17] . ...
Article
Full-text available
Η παρούσα εργασία αφορά την βιβλιογραφική ανασκόπηση του προγράμματος ‘OSTEODENT’ κατά το οποίο εξετάζονται οι πιθανότητες ένδειξης οστεοπενίας και κατ’ επέκτασιν οστεοπόρωσης, μέσω οδοντιατρικών ακτινογραφιών. Σκοπός της εργασίας είναι να συγκεντρώσει τα μέχρι στιγμής υπάρχοντα δεδομένα και συμπεράσματα που προέκυψαν από το εν λόγω πρόγραμμα, παραθέτοντας επιγραμματικά τον τρόπο επεξεργασίας των δεδομένων που συλλέχθηκαν. Η βάση δεδομένων αποτελείται από οδοντιατρικές πανοραμικές ή και οπισθοφατνιακές ακτινογραφίες γυναικών ηλικίας 45-70 ετών, πάνω στις οποίες έχουν γίνει κάποιες μετρήσεις οστικής πυκνότητας, αλλά και από τις αντίστοιχες εμπεριστατωμένες μετρήσεις οστικής πυκνότητας των ιδίων ατόμων. Η σύγκριση των δεδομένων αυτών, είχε ως σκοπό την ανάδειξη της εγκυρότητας των πρώτων, στο βαθμό που υποδεικνύουν την ύπαρξη μεταβολικής νόσου των οστών. Τα στοιχεία επεξεργάστηκαν τέσσερα διαφορετικά πανεπιστημιακά ιδρύματα της Ευρώπης, χρησιμοποιώντας προ-υπάρχουσες κατηγοριοποιήσεις. Κατά την διαδικασία όμως προέκυψε η ανάγκη για νέα συστήματα καταγραφής αυτών των δεδομένων, καθώς και η σύζευξη κάποιων παραγόντων. Τα συμπεράσματα της κάθε έρευνας ξεχωριστά, αλλά και στο σύνολό τους, συγκλίνουν προς την περεταίρω αναγκαιότητα για μελέτη, καθώς οι ενδείξεις οστεοπενίας σε απλές ακτινογραφίες είναι ενθαρρυντικές, όχι όμως ακόμη διαγνωστικά έγκυρες.
... Of the 16 studies that evaluated correlation between tooth loss and systemic tion could be due to a relatively small sample size and inclusion of confounding factors such as age in the earlier studies.43,49,58 When corrected for confounding factors including age and smoking, a large group multicenter study showed compelling association between tooth loss with systemic BMD.63 Of note, tooth loss is also a clinical endpoint associated with multiple factors beyond periodontitis. ...
Article
Full-text available
Periodontitis and osteoporosis are prevalent inflammation‐associated skeletal disorders that pose significant public health challenges to our aging population. Both periodontitis and osteoporosis are bone disorders closely associated with inflammation and aging. There has been consistent intrigue on whether a systemic skeletal disease such as osteoporosis will amplify the alveolar bone loss in periodontitis. A survey of the literature published in the past 25 years indicates that systemic low bone mineral density (BMD) is associated with alveolar bone loss, while recent evidence also suggests a correlation between clinical attachment loss and other parameters of periodontitis. Inflammation and its influence on bone remodeling play critical roles in the pathogenesis of both osteoporosis and periodontitis and could serve as the central mechanistic link between these disorders. Enhanced cytokine production and elevated inflammatory response exacerbate osteoclastic bone resorption while inhibiting osteoblastic bone formation, resulting in a net bone loss. With aging, accumulation of oxidative stress and cellular senescence drive the progression of osteoporosis and exacerbation of periodontitis. Vitamin D deficiency and smoking are shared risk factors and may mediate the connection between osteoporosis and periodontitis, through increasing oxidative stress and impairing host response to inflammation. With the connection between systemic and localized bone loss in mind, routine dental exams and intraoral radiographs may serve as a low‐cost screening tool for low systemic BMD and increased fracture risk. Conversely, patients with fracture risk beyond the intervention threshold are at greater risk for developing severe periodontitis and undergo tooth loss. Various Food and Drug Administration‐approved therapies for osteoporosis have shown promising results for treating periodontitis. Understanding the molecular mechanisms underlying their connection sheds light on potential therapeutic strategies that may facilitate co‐management of systemic and localized bone loss.
... High body mass index is protective, whereas a history of fracture, glaucoma, smoking, negative attitude, and the presence of medical conditions may increase the prevalence of osteoporotic fractures [31]. Tooth loss is associated with cortical erosion [32] and osteoporosis [33,34]. ...
Article
Full-text available
Objective To investigate the association between mandibular cortex parameters and fracture in a group of 286 men and women, 79–80 years of age. Study design In a cross-sectional study, the mandibular cortex was evaluated with Klemetti’s index for cortical erosion. The cortical thickness was measured with a ruler adjusting for the magnification factor. The odds ratio (OR) for fracture when having a severely eroded cortex or a cortex thickness < 3 mm was calculated. Results A normal cortex was found in 65% of men, whereas only 7% had a severely eroded cortex. The OR for severely eroded cortex as fracture risk predictor was significant (2.32; 95% CI 1.3–4.2), also when the female group was evaluated separately. A significant difference was found between the mean thickness for men (3.96 mm) and women (2.92 mm), respectively. The OR for cortical thickness < 3 mm was significant (2.00; 95% CI 1.1–3.6) in the total group, but not when men and women were evaluated separately. Conclusions Among old women, the cortical parameters were significantly associated with prevalent fracture. In old men, other circumstances may be more important.
... During the studies done to investigate relationship between osteoporosis and missing teeth, it was seen different results. In Nicopoulou-Karayianni, 16 Bollen et al., 17 studies there was not also significant relationship between osteoporosis and tooth loss and this finding was in agreement with this work. ...
Article
Introduction: Little is known about whether substances inducing tissue protein degeneration in the oral cavity are associated with the number of teeth present in postmenopausal women. We sought to investigate the association of urinary pentosidine and serum homocysteine levels with the number of teeth and subsequent tooth loss in Japanese postmenopausal women. Materials and methods: Among participants in the Nagano Cohort Study, 785 postmenopausal women (mean age, 68.1 years) participated in the present study. The number of teeth was re-counted at the time of follow-up in 610 women. Poisson regression analysis was used to investigate differences in the number of teeth among quartiles of pentosidine or homocysteine, adjusting for covariates that correlated with the number of teeth. A Cox proportional hazard model was used to evaluate the association of subsequent tooth loss with pentosidine or homocysteine levels. Results: Pentosidine quartiles were not associated with the number of teeth at baseline. Participants in the highest homocysteine quartile had significantly fewer teeth at baseline than those in the third and lowest quartiles (p < 0.001 for both). Those in the second quartile had fewer teeth than those in the third (p = 0.001) and lowest (p < 0.001) quartiles. An increased risk of tooth loss during follow-up was significantly associated with higher urinary pentosidine (hazard ratio = 1.073 for 10 pmol/mgCre; p = 0.001). Conclusion: Postmenopausal women with higher homocysteine levels had fewer teeth at baseline. A higher pentosidine concentration increased the risk of subsequent tooth loss. High pentosidine or homocysteine concentrations may be associated with tooth loss in postmenopausal women.
Article
Full-text available
We empirically test existing theories on the provision of public goods, in particular air quality, using data on sulfur dioxide (SO2) concentrations from the Global Environment Monitoring Projects for 107 cities in 42 countries from 1971 to 1996. The results are as follows: First, we provide additional support for the claim that the degree of democracy has an independent positive effect on air quality. Second, we find that among democracies, presidential systems are more conducive to air quality than parliamentary ones. Third, in testing competing claims about the effect of interest groups on public goods provision in democracies we establish that labor union strength contributes to lower environmental quality, whereas the strength of green parties has the opposite effect.
Book
A guide to using S environments to perform statistical analyses providing both an introduction to the use of S and a course in modern statistical methods. The emphasis is on presenting practical problems and full analyses of real data sets.
Chapter
Reference data have a major impact on our daily lives in many ways. Clinical measurements of various sorts are used to assess our health: are we too heavy; is our blood pressure too high; is our blood cholesterol normal? All these judgements are based on comparison of the measurement to accepted reference data sets.
Article
Purpose: Oral bone and tooth loss are correlated with bone loss at nonoral sites. Calcium and vitamin D supplementation slow the rate of bone loss from various skeletal sites, but it is not known if intake of these nutrients affects oral bone and, in turn, tooth retention. Subjects and methods: Tooth loss was examined in 145 healthy subjects aged 65 years and older who completed a 3-year, randomized, placebo-controlled trial of the effect of calcium and vitamin D supplementation on bone loss from the hip, as well as a 2-year follow-up study after discontinuation of study supplements. Teeth were counted at 18 months and 5 years. A comprehensive oral examination at 5 years included assessment of caries, oral hygiene, and periodontal disease. The odds ratio (OR) and 95% confidence interval (CI) of tooth loss were estimated by stepwise multivariate logistic regression. Initial age (mean +/- SD) of subjects was 71 +/- 5 years, and the number of teeth remaining was 22 +/- 7. Results: During the randomized trial, 11 of the 82 subjects (13%) taking supplements and 17 of the 63 subjects (27%) taking placebo lost one or more teeth (OR = 0.4; 95% CI: 0.2 to 0.9). During the 2-year follow-up period, 31 of the 77 subjects (40%) with total calcium intake of at least 1000 mg per day lost one or more teeth compared with 40 of the 68 subjects (59%) who consumed less (OR = 0.5; 95% CI: 0.2 to 0.9). Conclusion: These findings suggest that intake levels of calcium and vitamin D aimed at preventing osteoporosis have a beneficial effect on tooth retention.
Article
It has been suggested that periodontitis and systemic bone mass might be related. In order to evaluate this possible relationship, we performed an intra-oral examination and measured lumbar bone mineral density (lumbar BMD) and metacarpal cortical thickness (MCT) in 286 female volunteers between 46 and 55 years of age. In addition, the alveolar bone height was measured on bite wing radiographs of the dentate subjects. Of the subject, n = 60 (21%) were edentulous. Compared to the dentate subjects, the lumbar BMD and MCT of the edentulous women were not significantly different. In the dentate subjects, no significant correlation was observed between the clinical parameters of periodontitis (mean probing depth, occurrence of bleeding after probing and number of missing teeth) and the bone mass parameters (lumbar BMD and MCT); nor was a significant relation observed between the bone mass measurements and alveolar bone height. We therefore suggest that systemic bone mass is not an important factor in the pathogenesis of periodontitis.
Article
This study compared dental findings in normal and osteoporotic women and evaluated the variables that best separated the two groups. The osteoporotic group had less mandibular bone mass and density and a thinner cortex at the gonion than the normal group. The osteoporotic group also had a greater percentage of subjects who were edentulous. In subjects who had natural teeth, there was greater tooth loss. No differences in periodontal measurements were found between osteoporotic and normal groups. Mandibular bone mass and the number of teeth were statistically most effective for separating the populations. However, considerable overlap was found between the osteoporotic and normal groups in all variables.