This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Mucci, L.A. Articles by Pedersen, N.L. Search for Related Content PubMed PubMed Citation Articles by Mucci, L.A. Articles by Pedersen, N.L. Social Bookmarking                         What's this?

Environmental and Heritable Factors in the Etiology of Oral Diseases—A Population-based Study of Swedish Twins

¹ Department of Epidemiology, Harvard School of Public Health, Boston, MA, USA;
² Channing Laboratory, Harvard Medical School/Brigham and Women’s Hospital, 181 Longwood Avenue, Boston, MA 02115, USA;
³ Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden;
⁴ Department of Oral Sciences-Center for Clinical Dental Research, University of Bergen, Norway;
⁵ Department of Oral Health Policy and Epidemiology, Harvard School of Dental Medicine, Boston, MA, USA; and
⁶ Department of Psychology, University of Southern California, Los Angeles, CA, USA;

Correspondence: ^* corresponding author, lmucci{at}hsph.harvard.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
A population-based twin study is a useful design for quantification of the effects of genes and environmental factors in disease etiology. We used data from 10,000 Swedish twin pairs to quantify genetic and environmental contributions to tooth loss and periodontal health. Oral health information was obtained from telephone interviews. Structural equation models measured the relative importance of genetic and environmental factors. Genetic factors contributed to 14% of variation in tooth loss among women, and 39% among men. Non-shared environmental factors accounted for one-quarter of risk; environmental factors shared by twins comprised the remainder. Heritability estimates of periodontal disease were 39% and 33% for women and men, respectively, while non-shared environmental factors accounted for the remaining variation. Heritability for both conditions varied as a function of age and smoking status. Analysis of data from this large, population-based study demonstrates a moderate role of genetic factors in oral diseases, and suggests potential gene-environment interactions.

Key Words: twin study • periodontal disease • tooth loss • heritability

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Periodontal disease is one of the most common dental diseases (Hugoson et al., 1998; Albandar et al., 1999). Bacterial agents have been confirmed as necessary causes of periodontal disease. However, bacteria alone are not sufficient causes, since their presence does not invariably result in disease. Additional co-factors are required that modify an individual’s susceptibility and affect disease course.

Recent studies have shed light on a possibly substantial role of genes in the etiology of periodontal disease, while age and smoking have been established as important risk factors. Understanding the relative contributions of genetic and environmental effects, and possible gene-environment interactions, is essential from both etiological and public health orientations. A population-based twin study is one of the best designs to quantify these effects.

To date, few twin studies of periodontal disease have been undertaken (Michalowicz et al., 1991, 2000; Corey et al., 1993), although the results are converging and suggest a substantial role of genetics. In the present study, we used data from the Swedish Twin Registry on over 10,000 twin pairs to estimate the effects of genetic and environmental factors on tooth loss and periodontal disease, and to quantify potential gene-environment interactions.

	MATERIALS & METHODS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Study Population
Data for the present study came from the Swedish Twin Registry (STR; www.meb.ki.se/twins), currently the largest population-based twin registry in the world, registering the more than 70,000 twin pairs born in Sweden since 1886. Between 1998 and 2002, all twins ages 42 yrs and above were invited to participate in Screening Across the Lifespan of Twins (SALT), a computer-assisted telephone interview collecting information on health outcomes, risk factors, and familial and demographic information.

Zygosity of twins was determined based on responses to the following question: "During childhood, were you and your twin partner as alike as ‘two peas in a pod’ or not more alike than siblings in general". If both members of a twin pair responded "as alike as two peas in a pod", then they were classified as monozygotic (MZ) (Pedersen et al., 2002). If both responded that they were not more alike than siblings in general, they were classified as dizygotic (DZ). If there was disconcordance between twins, or if one twin did not respond, zygosity was estimated through an algorithm based on additional questions (Lichtenstein et al., 2002).

Oral Health Measures
Questions on oral health were included in the SALT telephone interview. All twins were asked, "Do you have your own teeth?". Edentulism was defined as complete tooth loss. Participants who were not edentulous were also asked: "Have you been diagnosed by a dentist as having loose teeth or periodontal disease?" and "Do you have or have you ever had any teeth which are loose or move around?". If a twin responded yes to either, s/he was defined as having periodontal disease. Finally, twins were asked, "Do you have now, or have you ever had, problems with bleeding gums?". Gingivitis was defined based on ever having bleeding gums. The oral status of each twin was defined independent of his/her co-twin’s status.

Statistical Analysis
Three analytical approaches were undertaken to estimate the genetic and environmental contributions to oral disease. First, we calculated the relative risk of disease for persons whose co-twin had disease compared with those whose co-twin did not. We estimated Mantel-Haenszel relative risks, controlling for potential confounding by age, sex, and smoking, and stratified by zygosity. Relative risks above 1 suggest a familial component for the measure.

Second, we assessed twin similarity in oral disease status by calculating tetrachoric correlations using the PLCORR CL option in PROC FREQ in SAS (Version 8.2). Tetrachoric correlations are presented stratified by zygosity. The observed correlation among twin pairs is evidence of a familial component, while the extent to which correlations differ across MZ and DZ pairs provides evidence of a genetic effect.

Finally, we used structural equation models to quantify the relative importance of genetic and environmental factors, using Mx statistical software (Neale et al., 1999). In this method, variability in disease risk is partitioned into variance components for genes (a²), for shared environment to which both members of a twin pair reared together are exposed (c²), and individual-specific environmental effects (e²), which are exposures not necessarily shared by twins (Neale and Cardon, 1992). Models were fitted directly to contingency tables, i.e., disease present/absent in Twin 1 vs. Twin 2. Sex-limitation models were applied to incorporate data from all five types of twin pairs, and to estimate genetic and environmental effects separately for men and women. Genetic correlations were specified as 1.0 and 0.5 for MZ and DZ twins, respectively, since MZ twins share, in theory, 100% of their genes, while DZ twins share 50%. We used log-likelihood ratio tests to see whether coefficients were significantly different from zero. We computed goodness-of-fit statistics to assess the model fit. Furthermore, we used a full model, assessing all parameters (a², c², e²), to estimate heritability separately for twin pairs concordant for risk factors and concordant for no risk factor, and to assess whether heritability differed as a function of age or smoking.

We assessed the contributions of genetic and environmental factors separately for tooth loss, periodontal disease, and gingivitis, and twin pairs were included in the analyses for each oral disease for which both in the pair had available data. Twins could be considered concordant for some conditions and discordant on others. For example, if Twin A in a pair reported having both periodontal disease and gingivitis, while co-twin Twin B reported only gingivitis, the pair would be characterized as discordant on periodontal disease but concordant for gingivitis. Because twins edentulous at baseline were not asked about their periodontal disease history, pairs where one or both twins were edentulous were excluded from the analyses of periodontal disease and gingivitis.

The SALT study was approved by the ethical committee at the Karolinska Institutet, and all twins gave informed consent to participate.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Oral health information from SALT was available for 10,578 complete twin pairs ages 42 yrs and older: 1567 MZ and 2127 DZ female pairs, 1180 MZ and 1696 DZ male pairs, and 4008 unlike-sex DZ pairs. Among the twins, 6% were edentulous, 18% had ever had periodontal disease, and 24% had ever had gingivitis.

Mantel-Haenszel relative risks (RR_MH) were used to estimate the risk of an oral disease for persons whose co-twin had a disease, compared with those whose co-twin did not (Table 1). Persons whose co-twin was edentulous had a three-fold increased risk of edentulism (RR_MH = 2.9, 95% CI = 2.5–3.4) compared with those whose co-twin was not. The effect was stronger among MZ compared with DZ twins (RR_MH = 4.2 vs. RR_MH = 2.6, respectively). Having a twin who had periodontal disease increased one’s own risk of disease 1.5 times. The risk of periodontal disease or gingivitis was increased among persons whose co-twin had ever had disease, and was higher among MZ vs. DZ pairs.

We examined whether estimates of heritability varied as a function of twin characteristics (Table 4). The proportion of variance in edentulism due to genetic effects was two-fold higher among twins older than age 65 yrs, compared with younger twins, while heritability estimates did not differ by smoking status. For periodontal disease, heritability varied by age group and smoking status, with little contribution among older twins or those who never smoked.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

These findings are in line with data from earlier studies. In two small twin studies (110 and 117 pairs, respectively), Michalowicz and colleagues (Michalowicz et al., 1991, 2000) assessed periodontal measures clinically, and found that a substantial proportion (40–50%) of the variance in risk was attributed to genes. In a larger study of 4000 twin pairs with self-reported oral health (Corey et al., 1993), concordance for periodontal disease was higher among MZ (38%) compared with DZ (16%) twins, lending further support for a significant genetic component.

While twin studies demonstrate a substantial contribution of genetic factors, there is limited evidence identifying specific genes. The most consistent findings involve a composite genotype of Interleukin-1A (IL-1A) and IL-1B polymorphisms associated with a higher risk of periodontal disease (Diehl et al., 1999; Kornman et al., 1999). There is suggestive evidence for additional candidate genes, including TNF-alpha and immunoglobulin G Fc receptor, although the results have thus far been inconclusive (Kornman and Duff, 2001).

Shared environmental factors represent the familial experiences and habits common within twin pairs. In our data, shared-environment did not play a significant role in periodontal disease. Similarly, Michalowicz and co-workers (2000) found no evidence of shared environmental effect. Structural equation models in twin studies may have limited power to detect small effects of shared environment, however (Christian et al., 1995; Ramakrishnan et al., 1996). We observed that shared environment accounted for about 20% of the variability in periodontal disease among non-smokers. Future studies should not completely exclude the possibility that shared environmental factors could play a modest role in the presence of oral diseases among subgroups.

Non-shared environmental factors accounted for the majority of variability in periodontal disease in our study. Epidemiological studies support this finding, and have demonstrated a strong association between cigarette smoking and periodontal health (Tomar and Asma, 2000). Given the strength of the relationship and the prevalence of this risk factor, smoking is likely to be a major non-shared environmental risk factor that is potentially modifiable. While not all environmental factors are modifiable, analysis of our data suggests that a substantial proportion of periodontal disease is potentially preventable through factors such as smoking.

Twins who reported being edentulous at baseline were not queried about periodontal disease history, and thus pairs in which one or both were edentulous would not be included in the analyses of periodontal disease. We would expect our estimates of heritability for periodontal disease to be robust, however, since the proportion of pairs excluded from the analyses was low, and because it is likely that the relative genetic and environmental contributions to periodontal disease are similar in pairs with and pairs without tooth loss.

We could find no twin studies of tooth loss. In our data, we found strong familial effects, both inherited and non-hereditary. In contrast to periodontal disease, shared environmental effects contributed to variability in risk of tooth loss, particularly among women. Non-shared environment made a much smaller contribution. While periodontal disease is an important cause of tooth loss, other oral diseases, notably dental caries, contribute to edentulism. Also, there may be influences which affect the propensity to lose one’s teeth, regardless of the presence of periodontal disease. If our data are substantiated in future studies, these findings suggest that the risk of tooth loss is predicted, in large part, early in life, through genetic predisposition and early-life exposures, an observation substantiated in an earlier study among Swedish twins that demonstrated familial influences during childhood on tooth loss (Mucci et al., 2004).

Gender differences in the risk of oral diseases have been consistently observed (Albandar et al., 1999). Such disparities may indicate differences in the prevalence of risk factors between men and women. Conversely, they could suggest biological differences in response to a genetic predisposition. Analyses of large samples of unlike-sex twins can address this issue. Using sex-limitation models, we estimated gender-specific effects of heritable and environmental factors. While we noted only minimal differences in genetic effects on periodontal disease, heritability estimates for edentulism were substantially greater among men than women. It is unlikely that differences in the prevalence of risk factors between men and women accounted for the differences in heritability estimates, since the gender-specific patterns were consistent for smokers and non-smokers, and across age groups. Instead, the higher estimates of heritability among men could reflect different biological responses to genetic susceptibility to tooth loss. In contrast, it is possible that the underlying causes of tooth loss are different among men and women. Finally, given the larger role of shared environment for women than men, it may be that the future oral care patterns of women are, to a greater extent, influenced by the family environment than they are for men. For example, the likelihood that a woman has good oral hygiene practices is strongly related to her exposure to hygiene practices in childhood.

Given changes in the social structure in Sweden and the prevalence of periodontal risk factors, the relative contributions of genetic factors may differ across birth cohorts. Indeed, analysis of our data suggests substantial age differences. Among the older cohort, genetics was an important predictor of edentulism risk, whereas heritability for periodontal disease was almost zero. Our study was cross-sectional, which prevented us from disentangling age from birth cohort effects. Additional studies of twin cohorts with longitudinal data are needed to address this question.

The effect of smoking on periodontal disease may be mediated through an inflammatory response. Conversely, genetic influences for the propensity to smoke may account for some of the heritability for periodontal disease. Thus, the genetic influence on periodontal disease may differ for smokers and non-smokers. Data from our study support this, with a stronger genetic effect noted among twins where both smoked. This finding could be evidence of a gene-environment interaction. No such difference in heritability was noted for edentulism.

This study is subject to the possible limitations of twin studies. First, the validity of results from the structural equation model depends on the assumption of equal environments and random mating. ‘Equal environments’ implies that MZ and DZ twins are equally correlated in their exposure to environmental events of etiologic importance from birth through adulthood for the trait under study. This assumption appears to hold for a variety of conditions (Kendler et al., 1993; Kendler and Gardner, 1998). Assortive, or non-random, mating can inflate the correlation among DZ twins and bias estimates of heritability toward the null (Plomin et al., 2001). To the extent that environmental characteristics are similar between mates, there could be phenotypic similarities with respect to oral diseases. However, correlation between parents of the twins on periodontal measures is likely to be low.

The data collected through SALT were based on self-report, and may be subject to some misclassification. Validation studies undertaken in the STR have demonstrated that 98% of MZ pairs and 94% of DZ pairs were correctly classified based on self-report (Pedersen et al., 2002). Self-reported periodontal measures have created considerable debate among oral health professionals, although few studies have examined their validity. Self-report of tooth loss appears to have high validity compared with clinical assessment (Könönen et al., 1986; Douglass et al., 1991). Among health professionals, validity of self-reported periodontal disease was high, particularly for those with severe disease (Joshipura et al., 1996, 2002). In a community sample, agreement between self-report of periodontal pockets and clinical exam was relatively high (Buhlin et al., 2002), while the correlation for tooth mobility was lower. Within the STR, we find supportive evidence for minimal misclassification of oral health measures, since edentulism and periodontal disease were associated with their established risk factors. Also, data from a sample of twins re-contacted 2 wks after initial interview showed excellent reliability for edentulism (kappa = 86%) and periodontal disease (89%). If misclassification of outcome exists, it is likely to be non-differential across twin pairs. Moreover, the likelihood of undiagnosed oral disease would result in an underestimate, rather than an overestimate, of the prevalence of these conditions. Such misclassification would likely underestimate the contribution of heritability, by making the MZ and DZ twin pairs appear more similar. Thus, our estimates of the contribution of genetic factors in the etiology of oral disease may be conservative.

In conclusion, we found that genetic factors play an important role in liability to periodontal disease, gingivitis, and edentulism. Shared environmental factors contribute substantially to edentulism, while non-shared environment effects impart the greatest proportion of variance of periodontal disease. The contributions of genetic and environmental effects are modified to some extent by gender, age or birth cohort, and smoking.

	ACKNOWLEDGMENTS

 
The authors express their gratitude to the participants of the Swedish Twin Registry study, without whom this study would not have been possible. L.A. Mucci was supported by training grant number 2T32 DE07151 through the National Institute of Dental and Craniofacial Research at the NIH. Support for this project came from a STINT (Swedish Foundation for International Cooperation in Research and Higher Education) grant, supporting collaboration between Karolinska Institutet and Harvard University. SALT screening was supported by grants from the Swedish Council for the Planning and Coordination of Research (FRN) and from the US National Institute on Aging (AG 08724). The Swedish Twin Registry is supported by grants from the Swedish Department of Higher Education, the Swedish Research Council, and ASTRAZeneca. The authors thank Ms. Alexandra Ash for her editorial assistance in manuscript preparation.

Received for publication July 30, 2004. Revision received April 27, 2005. Accepted for publication June 18, 2005.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 

• Albandar JM, Brunelle JA, Kingman A (1999). Destructive periodontal disease in adults 30 years of age and older in the United States, 1988–1994. J Periodontol 70:13–29.[CrossRef][Medline] [Order article via Infotrieve]
• Buhlin K, Gustafsson A, Andersson K, Hakansson J, Klinge B (2002). Validity and limitations of self-reported periodontal health. Community Dent Oral Epidemiol 30:431–437.[CrossRef][Medline] [Order article via Infotrieve]
• Christian JC, Norton JA Jr, Sorbel J, Williams CJ (1995). Comparison of analysis of variance and maximum likelihood based path analysis of twin data: partitioning genetic and environmental sources of covariance. Genet Epidemiol 12:27–35.[Medline] [Order article via Infotrieve]
• Corey LA, Nance WE, Hofstede P, Schenkein HA (1993). Self-reported periodontal disease in a Virginia twin population. J Periodontol 64:1205–1208.[Medline] [Order article via Infotrieve]
• Diehl SR, Wang Y, Brooks CN, Burmeister JA, Califano JV, Wang S, et al. (1999). Linkage disequilibrium of interleukin-1 genetic polymorphisms with early-onset periodontitis. J Periodontol 70:418–430.[CrossRef][Medline] [Order article via Infotrieve]
• Douglass CW, Berlin J, Tennstedt S (1991). The validity of self-reported oral health status in the elderly. J Public Health Dent 51:220–222.[Medline] [Order article via Infotrieve]
• Hugoson A, Norderyd O, Slotte C, Thorstensson H (1998). Distribution of periodontal disease in a Swedish adult population 1973, 1983 and 1993. Clin Periodontol 25:542–548.
• Joshipura KJ, Douglass CW, Garcia RI, Valachovic R, Willett WC (1996). Validity of a self-reported periodontal disease measure. J Public Health Dent 56:205–212.[Medline] [Order article via Infotrieve]
• Joshipura KJ, Pitiphat W, Douglass CW (2002). Validation of self-reported periodontal measures among health professionals. J Public Health Dent 62:115–121.[Medline] [Order article via Infotrieve]
• Kendler KS, Gardner CO Jr (1998). Twin studies of adult psychiatric and substance dependence disorders: are they biased by differences in the environmental experiences of monozygotic and dizygotic twins in childhood and adolescence? Psychol Med 28:625–633.[CrossRef][Medline] [Order article via Infotrieve]
• Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ (1993). A test of the equal-environment assumption in twin studies of psychiatric illness. Behav Genet 23:21–27.[CrossRef][Medline] [Order article via Infotrieve]
• Könönen M, Lipasti J, Murtomaa H (1986). Comparison of dental information obtained from self-examination and clinical examination. Community Dent Oral Epidemiol 14:258–260.[Medline] [Order article via Infotrieve]
• Kornman KS, Duff GW (2001). Candidate genes as potential links between periodontal and cardiovascular diseases. Ann Periodontol 6:48–57.[CrossRef][Medline] [Order article via Infotrieve]
• Kornman KS, Pankow J, Offenbacher S, Beck J, di Giovine F, Duff GW (1999). Interleukin-1 genotypes and the association between periodontitis and cardiovascular disease. J Periodontal Res 34:353–357.[CrossRef][Medline] [Order article via Infotrieve]
• Lichtenstein P, De Faire U, Floderus B, Svartengren M, Svedberg P, Pedersen NL (2002). The Swedish Twin Registry: a unique resource for clinical, epidemiological and genetic studies. J Intern Med 252:184–205.[CrossRef][Medline] [Order article via Infotrieve]
• Michalowicz BS, Aeppli D, Virag JG, Klump DG, Hinrichs JE, Segal NL, et al. (1991). Periodontal findings in adult twins. J Periodontol 62:293–299.[Medline] [Order article via Infotrieve]
• Michalowicz BS, Diehl SR, Gunsolley JC, Sparks BS, Brooks CN, Koertge TE, et al. (2000). Evidence of a substantial genetic basis for risk of adult periodontitis. J Periodontol 71:1699–1707.[CrossRef][Medline] [Order article via Infotrieve]
• Mucci LA, Hsieh CC, Williams PL, Dickman PW, Bjorkman L, Pedersen NL (2004). Birth order, sibship size, and housing density in relation to tooth loss and periodontal disease: a cohort study among Swedish twins. Am J Epidemiol 159:499–506.[Abstract/Free Full Text]
• Neale MC, Cardon LR (1992). Methodology for genetic studies of twins and families. Dordrecht, The Netherlands: Kluwer Academic Publishers.
• Neale MC, Boker S, Xie G, Maes H (1999). Mx: statistical modeling. 5th ed. Richmond, VA: Medical College of Virginia, Department of Psychiatry.
• Pedersen NL, Lichtenstein P, Svedberg P (2002). The Swedish Twin Registry in the third millennium. Twin Res 5:427–432.[CrossRef][Medline] [Order article via Infotrieve]
• Plomin R, DeFries JC, McClearn GE, McGuffin P (2001). In: Behavioral genetics. New York: Worth Publishers, pp. 246.
• Ramakrishnan V, Meyer JM, Goldberg J, Henderson WG (1996). Univariate analysis of dichotomous or ordinal data from twin pairs: a simulation study comparing structural equation modeling and logistic regression. Genet Epidemiol 13:79–90.[Medline] [Order article via Infotrieve]
• Tomar SL, Asma S (2000). Smoking-attributable periodontitis in the United States: findings from NHANES III. National Health and Nutrition Examination Survey. J Periodontol 71:743–751.[CrossRef][Medline] [Order article via Infotrieve]

Journal of Dental Research, Vol. 84, No. 9, 800-805 (2005)
DOI: 10.1177/154405910508400904


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				L. A. Mucci, C.-c. Hsieh, P. L. Williams, M. Arora, H.-O. Adami, U. de Faire, C. W. Douglass, and N. L. Pedersen Do Genetic Factors Explain the Association Between Poor Oral Health and Cardiovascular Disease? A Prospective Study Among Swedish Twins Am. J. Epidemiol., September 1, 2009; 170(5): 615 - 621. [Abstract] [Full Text] [PDF]

				A.-L. Ostberg, M. Nyholm, B. Gullberg, L. Rastam, and U. Lindblad Tooth loss and obesity in a defined Swedish population Scand J Public Health, June 1, 2009; 37(4): 427 - 433. [Abstract] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Mucci, L.A. Articles by Pedersen, N.L. Search for Related Content PubMed PubMed Citation Articles by Mucci, L.A. Articles by Pedersen, N.L. Social Bookmarking                         What's this?