| Sign In to gain access to subscriptions and/or personal tools. |
Oral Symptoms Predict Mortality: a Prospective Study in Japan
1 Department of Work Systems and Health, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807-8555, Japan; Correspondence: * corresponding author, r-ochide{at}med.uoeh-u.ac.jp
Several studies have reported positive associations between oral infections and systemic diseases. The purpose of the present study was to evaluate the effects of oral symptoms on mortality from cardiovascular disease (CVD) and pneumonia. Using data from a cohort study in Japan, we analyzed 4,139 individuals aged 40–79 years. The baseline questionnaire included the following items related to oral symptoms: ‘sensitive teeth’, ‘difficulty in chewing tough food substances’, ‘bleeding gums’, and ‘mouth feels sticky’. We used the Cox proportional hazard model to estimate hazard ratios (HRs) and 95% confidence intervals (95%CIs) for mortality, after adjustments for lifestyle, socio-economic factors, and history of diseases. Persons complaining that their ‘mouth feels sticky’ had a two-fold higher risk of pneumonia (HR = 2.1; 95%CI, 1.2–3.6), while those complaining of ‘sensitive teeth’ had a lower risk of CVD (HR = 0.4; 95%CI, 0.2–0.9). Some oral symptoms may be predictors of mortality from pneumonia and CVD.
Key Words: oral health • cardiovascular disease • pneumonia • cohort study
Over the last decade, several studies have suggested that the state of the oral cavity may be linked to systemic diseases. Epidemiologic studies have reported associations between periodontal diseases and the risk of cardiovascular disease (CVD). A meta-analysis indicated that periodontal infection increases the risks of coronary heart disease and cerebrovascular disease (Janket et al., 2003; Khader et al., 2004). There is reasonable evidence that infectious and inflammatory processes play roles in the promotion of CVD (Amar and Han, 2003). Oral microbiological infections may affect a person’s general health status. Indeed, a recent randomized controlled trial found that intensive treatment of periodontitis resulted in an improvement in endothelial function (Tonetti et al., 2007). Furthermore, oral bacteria have been implicated in the occurrence of pneumonia (Scannapieco, 1999). There is a growing body of evidence indicating an association between pneumonia and oral health (Azarpazhooh and Leake, 2006). These findings need to be interpreted with caution, however, due to the presence of confounding factors, such as age, smoking, and socioeconomic status (SES). The associations between various markers of oral health and mortality have been examined in recent longitudinal studies (Jansson et al., 2002; Ajwani et al., 2003; Hamalainen et al., 2003, 2005; Tuominen et al., 2003; Abnet et al., 2005). Tooth loss was associated with increased rates of total death and death from upper gastrointestinal cancer, heart diseases, and stroke (Abnet et al., 2005). Although smoking was strongly correlated with both tooth loss and death, smoking as a confounding factor may not have contributed to the results (Abnet et al., 2005). In a cohort of community-dwelling people aged 18–65 yrs, poor oral health was associated with an increased mortality risk, even when deaths caused by CVD were eliminated from the analyses (Jansson et al., 2002). Taken together, these observations suggest that the correlations cannot be explained by the identified confounding factors. Self-reporting is an efficient and accepted means of assessing many diseases. A single item of a general self-rated health questionnaire is a powerful predictor of mortality, even after adjustment for key covariates, such as functional status, depression, and co-morbidity (DeSalvo et al., 2006). Self-rated oral health has a unique role in people’s perception of their overall health, but is not yet fully captured by self-rated health questionnaires (Benyamini et al., 2004). Oral symptoms are subjective perceptions of oral health that are derived from oral diseases and disorders, such as caries, periodontal disease, tooth loss, and xerostomia (dry mouth). To the best of our knowledge, no previous studies have focused on the associations between oral symptoms and mortality from some diseases, such as CVD and pneumonia. Since 1986, as a means of evaluating the relationships between lifestyle and health, a prospective cohort study (Miyako study) has been conducted in a general population derived from four areas of Fukuoka Prefecture, located in southwestern Japan. The Miyako study is part of a large-scale population-based cohort study, the Japan Collaborative Cohort Study (JACC Study) for the Evaluation of Cancer Risk, sponsored by the Ministry of Education, Science, Sports and Culture of Japan (Monbusho) (Tamakoshi et al., 2005). The baseline questionnaire of the Miyako study included the following items related to oral symptoms: ‘sensitive teeth’, ‘difficulty in chewing tough food substances’, ‘bleeding gums’ and ‘mouth feels sticky’. The present study prospectively examined the effects of oral symptoms on mortality from all causes, CVD, and pneumonia. The baseline lifestyle, SES information, and history of diseases for each person allowed us to adjust for potential confounding factors.
In several published papers regarding the Miyako study, the protocol and follow-up used in the present study have been described in detail (Utoguchi et al., 1997; Mizoue et al., 2000; Fujino et al., 2001, 2002; Ngoan et al., 2002; Pham et al., 2006). Briefly, all inhabitants aged 30–79 yrs in Town A and Village B, and selected districts of City C and Town D (15,417 people in total), were invited to participate in a baseline survey between 1986 and 1989. The baseline data were collected by means of a comprehensive self-administered questionnaire that elicited socio-demographic characteristics, lifestyle-related factors (smoking, alcohol drinking, diet, etc.), and general and oral symptoms, etc. The response rate was 84.3% (n = 13,000). Informed consent was obtained at the group level after the study objective and the confidentiality of the data were explained to the community leaders. Individuals in the cohort were followed up so that the investigators could verify whether they had died or relocated, with the collaboration of each municipal office. For the deceased persons, the mortality data were coded based on the International Classification of Disease, Revision 9 (ICD 9), after permission was received from the Director-General of the Prime Minister’s Office (Ministry of Public Management, Home Affairs, Post and Telecommunications). Each disease was defined as follows: CVD, 410–414, 430–438; pneumonia, 481–486. The present study analyzed follow-up data to 31 December 2003 for City C and Town D, and follow-up data to 30 November 1999 for Village B. The baseline questionnaire for Town A did not include items regarding oral symptoms, and this area was therefore excluded from the present analysis. Of the remaining 8000 individuals, we excluded: (1) those under 40 yrs of age (n = 1742), (2) those who were followed up for < 12 mos (n = 130), and (3) those whose information regarding lifestyle or SES was inadequate (n = 1989). The data for the remaining 2057 men and 2082 women were analyzed in the present study. If participants had complained of each oral symptom—i.e., ‘sensitive teeth’, ‘difficulty in chewing tough food substances’, ‘bleeding gums’, and ‘mouth feels sticky’—they were checked for the corresponding items. The responses were categorized into binary "yes" or "no" groups. We used the Cox proportional hazard model to estimate the hazard ratio (HR) and 95% confidence interval (95%CI) of each oral symptom for death from all causes, CVD, and pneumonia. To estimate multivariate HRs, we considered the following variables as potential confounders and included them in the model: age (yrs); sex; smoking status (current smoker, past smoker, non-smoker); alcohol consumption (daily drinker, > 1 time per mo, seldom drinker, or non-drinker); exercise (> 2 times per wk, < 2 times per wk); body mass index (BMI; > 25 kg/m2, < 25 kg/m2); educational background (attended school beyond 18 yrs of age, left school before 18 yrs of age); marital status (married, non-married); and history of myocardial infarction, angina pectoris, stroke, hypertension, or diabetes. All calculations were performed with the Statistical Analysis System (SAS), version 8.02.
During a mean follow-up period of 13.7 yrs, 715 people (17.3%) died and 309 (7.5%) had moved out of the study areas. Specifically, there were 110 deaths from CVD (70 men; 40 women) and 71 deaths from pneumonia (46 men; 25 women).
The baseline characteristics of the study individuals were summarized by sex (Table 1
The multivariate HR for the ‘difficulty in chewing tough food substances’ symptom for overall mortality showed statistical significance (HR = 1.2; 95%CI, 1.0–1.4; p = 0.025) (Table 2  ). There were no statistically significant associations between death due to all causes and other oral symptoms.
Persons who had the ‘sensitive teeth’ symptom had a lower risk of death due to CVD, and their decreased risk showed statistical significance (HR = 0.4; 95%CI, 0.2–0.9; p = 0.023) (Table 3 ). There were no clear associations between death due to CVD and other oral symptoms.
Persons with the ‘difficulty in chewing tough food substances’ symptom showed a slightly higher risk of death due to pneumonia than those without this symptom, although it was not statistically significant (HR = 1.4; 95%CI, 0.9–2.3; p = 0.157) (Table 4 ). Persons who had the ‘mouth feels sticky’ symptom had a two-fold higher risk of death due to pneumonia (HR = 2.1; 95%CI, 1.2–3.6; p = 0.007) (Table 4). No such associations were found for the ‘sensitive teeth’ and ‘bleeding gums’ symptoms.
The present study prospectively investigated the effects of oral symptoms on mortality due to all causes, CVD, and pneumonia, using data from a community-based cohort study in Japan. As expected, some oral symptoms were independently associated with mortality due to some diseases, even after adjustments for key covariates, such as lifestyle and SES factors. The main finding of the present study is that the ‘mouth feels sticky’ symptom is an independent predictor of death due to pneumonia. The specificities of these associations strengthen the hypothesis. The mechanism behind poor dental hygiene and increased risk of pneumonia has been clarified (Scannapieco, 1999). The pathogenesis of bacterial pneumonia begins with colonization of oropharyngeal surfaces by potential respiratory pathogens. The bacteria can be aspirated into the lower airway, where they may cause a lower respiratory infection, even in normal healthy individuals (Ozagar et al., 2000). The ‘mouth feels sticky’ symptom may be related to xerostomia (dry mouth), which is an implicit consequence of the side-effects of therapeutic drugs. There is a longitudinal study of the links between medication exposure and xerostomia (Thomson et al., 2006). In that study, recent exposure to diabetes or daily aspirin was strongly associated with the incidence of xerostomia. Our results indicate that people with the ‘mouth feels sticky’ symptom showed a two-fold higher risk of death from pneumonia, even after adjustment for a history of diabetes at baseline. In addition, we found that the ‘difficulty in chewing tough food substances’ symptom was slightly associated with an increased risk of death due to pneumonia, although it was not statistically significant. A relationship between hyposalivation and masticatory performance has been observed in Japanese people over 60 yrs of age (Ikebe et al., 2006). Nutrient status is one of the indicators of an adverse prognosis of pneumonia (Janssens, 2005). Our findings revealed that the ‘difficulty in chewing tough food substances’ symptom was significantly associated with overall mortality. Tooth loss and unstable dentures have negative impacts on masticatory ability and nutrient intake (Sheiham et al., 1999). Poor diet and impaired food choice have been reported to be associated with declining numbers of teeth and increasing age (Daly et al., 2003). A certain impairment of masticatory function that interferes with eating an adequate diet may lead to increased mortality. Although several studies have found that periodontal disease is significantly correlated with CVD, it is possible that these correlations may be the result of unsolved confounding effects. A recent meta-analysis reported that the estimated relative risks of coronary heart disease and cerebrovascular disease among people with periodontitis, compared with healthy individuals, were 1.15 (95%CI, 1.06–1.25) and 1.13 (95%CI, 1.01–1.27), respectively (Khader et al., 2004). At the time of the baseline survey used in our study, the prevalence of periodontitis was approximately 30–50% among individuals aged between 40 and 79 yrs (Health Service Bureau, 1989). However, the lack of relationships between the examined symptoms and the mortality due to CVD in the present study does not seem to support the results of previous studies. A larger number of missing teeth at baseline was reported to be associated with higher risks for total death and death from upper gastrointestinal cancer, heart diseases, and stroke in a Chinese population-based cohort (Abnet et al., 2005). In our study, persons showed a lower risk of CVD linked to the ‘sensitive teeth’ symptom. We hypothesized that those who rated the ‘sensitive teeth’ symptom had enough remaining teeth to feel pain. In other words, the ‘sensitive teeth’ symptom may be a surrogate marker for a greater number of remaining teeth, which is related to a decreased risk of mortality from CVD. However, the mechanism that underlies the interpretation of the association between tooth loss and mortality is unclear. The strength of the present study is that it represents a prospective cohort study with adjustment for possible confounders. In particular, SES factors are known to play important roles in relation to oral conditions and mortality. The utilization of medical and dental care may differ among different SES levels. Awareness and perceptions of oral conditions may also differ. Our findings indicate that some oral conditions were still associated with mortality after adjustments for both lifestyle and SES factors (educational level, marital status). In addition, we controlled for other confounding factors, such as history of myocardial infarction, angina pectoris, stroke, hypertension, or diabetes. However, the results may need to be interpreted with a degree of caution. We cannot conclude a causal relationship between oral diseases and mortality in the present study, since clinical measures, such as indices of periodontal disease and tooth loss, were not used. Oral diseases and disorders lead to functional limitations, pain, and discomfort, which must be reported by the individuals themselves. Hence, oral symptoms are multifactorial, and do not specify one disease. If oral symptoms are considered as a comprehensive measure of oral conditions, the data are more readily available than clinical measures. Some limitations need to be considered. First, information on oral symptoms was collected only at baseline. The deterioration of oral conditions, such as periodontal disease and tooth loss, becomes worse between the ages of 40 and 79 yrs (Health Service Bureau, 1989), and this could result in potential underestimation of the impact of oral symptoms. Second, there were non-respondents for items of lifestyle and SES factors, especially smoking and drinking habits in women (percentages of missing data for women: smoking, 22%; alcohol drinking, 21%). We included the persons with complete data in the present analysis. The results were consistent when the missing information for covariate data was included in the multivariate analyses as the ‘unknown’ category. Finally, the history of disease was based on self-reporting and may be subject to misclassification. In summary, our study in middle-aged and elderly individuals has revealed that some oral symptoms can predict mortality. Numerous studies regarding the effects of oral conditions on systemic health have been accumulated to date. To the best of our knowledge, the present study is the first report on the associations of oral symptoms with mortality in a longitudinal study. Further research is required to clarify whether oral health promotion reduces the risk of systemic diseases.
The work has been supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan (Monbusho) (Nos. 61010076, 62010074, 63010074, 1010068, 2151065, 3151064, 4151063, 5151069, 6279102, and 11181101), as part of the JACC study. We thank all the study participants, the public health authorities in the municipal offices, and the public staff at Miyako Health Center and Munakata Health Center for their cooperation. We also wish to thank Ms. Yukiko Fujino (née Takano), Ms. Yuko Uemura, and Ms. Yoko Wada for helping with the project. Received for publication March 12, 2007. Revision received January 17, 2008. Accepted for publication February 1, 2008.
Journal of Dental Research, Vol. 87, No. 5,
485-489 (2008) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
