Association Between Periodontitis and All-cause and Cancer Mortality

Retrospective Elderly Community Cohort Study

Ping-Chen Chung; Ta-Chien Chan


BMC Oral Health. 2020;20(168) 

In This Article


In this large elderly cohort, all-cause mortality and all-cancer mortality had positive association with periodontitis after adjusting for age, marital status, education level, sex and smoking status in Cox frailty models. Furthermore, in terms of specific cancers, lung and prostate cancer mortality were higher in the periodontitis group after adjusting for age and sex.

Bacterial biofilm invades the surrounding connective tissue of the gingiva, which may cause periodontitis. Direct or hematogenous spread of microorganisms increases the blood levels of inflammatory mediators, such as lipopolysaccharides and cytokines.[32,33] The bacteria of dental plaque triggering systemic reactions are thought to lead to malignant transformation in a variety of tissues.[34,35] Oral bacteria and inflammation may play a role in carcinogenesis.[36] The present findings show that dental plaque and gingival inflammation are associated with all-cancer mortality.[33] Furthermore, comorbidity and personal health behavior such as smoking status are important risk factors for periodontitis. Patients with periodontitis are more likely to have other systemic disease, an unhealthy lifestyle and low socioeconomic status, which also increases the mortality risk.[37]

In our findings, lung cancer and prostate cancer had positive association with periodontitis after adjusting for age and sex. Previous studies were less consistent in finding a relationship between lung cancer and periodontitis. Hujoel et al. in a US-population-based cohort in which periodontal status was defined by standard dental examination, after adjusting for age category and sex, found a positive association between lung cancer mortality and periodontitis (odds ratio = 1.97 (95%CI:1.21–3.22)).[7]

In another study, when comparing lung cancer mortality rates among participants with and without periodontitis defined by whether they had periodontitis treatment procedure codes, the results indicated no significant difference in crude (1.31 (95%CI: 0.92–1.86)) and adjusted mortality rate ratios (1.20 (95%CI: 0.81–1.80)) which were adjusted for calendar time, age, sex, socio-economic status, number of teeth, dental treatments, oral health indices, need of periodontal treatment, and diabetes.[9] Arora et al., in a Swedish Twin Registry longitudinal study, suggested a positive association between incidence of prostate cancer and periodontitis classified by self-report after adjusting for potential confounders including sex, age, education, employment, number of siblings, smoking status, smoking status of partner, alcohol status, diabetes, and body mass index (odds ratio = 1.47 (95%CI:1.04–2.07)).[38] Periodontal diseases can lead to tooth loss due to inflammatory conditions.[39] Hiraki et al., in a case control study, used tooth loss as an indicator, and a decreased number of teeth remaining was associated with a lower odds ratio for prostate cancer of 0.49 after adjusting for age, sex, smoking and drinking status, vegetable and fruit intake, BMI, and regular exercise.[40]

However, the associations between some specific cancers and periodontitis are not consistent with previous research, including esophageal, pancreatic, colorectal, liver and gallbladder cancer. One of the reasons might be the low number of esophageal cancer cases (n = 64), which made inference uncertain. The lack of an association between periodontitis and other cancers may be due to characteristics of the participants. In some studies, smoking appears to be a confounding factor in risk of cancer, especially for cancers strongly linked to tobacco use such as lung cancer.[7,10,41] Hujoel et al. pointed out that when limited to non-smokers, no association between periodontitis defined by standard dental examination and lung cancer mortality was identified after adjusting for age, age squared, race, poverty index, education, and vitamin A and C; however, when limited to smokers, periodontitis was shown to be significantly associated with lung cancer.[7] For pancreatic cancer, the hazard ratio was not found to be substantially different after adjusting for smoking.[10] Michaud et al. have suggested that smoking is not likely to account for the excess risks at cancer sites such as pancreatic and kidney cancers.[10]

Patients with diabetes have higher risk of developing periodontal disease. At the same time, periodontal disease may worsen the glucose control in diabetes patients. Several studies emphasize the impact of diabetes mellitus on subsequent risk of cancer.[42–44] In a Japanese hospital-based longitudinal study, a history of diabetes was associated with cancer risks for all sites for both males and females, controlling for age and potential confounders.[45] Patients with a past medical history of diabetes may have an increased incidence rate of specific cancers, notably pancreatic, liver, and colorectal cancer.[44] Previous studies usually took diabetes into account when exploring the relationship between periodontitis and mortality. One longitudinal population-based health survey showed a hazard ratio of 1.01 (95% CI: 1.002 to 1.01) for the association between the extent of clinical attachment level ≥ 3 mm and all-cause mortality after adjusting for age, sex, household income, years of education, body mass index, smoking, physical activity, and dental checkup. However, the findings did not indicate additive interaction of periodontal destruction and diabetes regarding all-cause mortality. Kebede et al. pointed out that despite the reciprocal relationship between periodontal destruction and diabetes, they may be independent risk factors for all-cause mortality.[46]

Previous studies have proposed a mechanism by which vitamin D decreases the risk of periodontal disease and decreases the risk of several cancers.[47] Individuals with less vitamin A and C are more likely to have periodontitis.[7] Individuals with less fruit and vegetable intake have increased risk of cancer of specific sites such as oral cancer and gastric cancer.[48] Less intake of fruit and vegetables may be due to tooth loss resulting from advanced periodontal disease.[39]

Our limitations include, first, the fact that the models were not adjusted for comorbidities such as respiratory disease, because the dataset didn't record respiratory function and respiratory diseases. Besides, physical activity was missing in 2005 and 2012, so we couldn't include this variable for analysis. Second, the outcome and the exposure could be affected by unmeasured confounding factors. And third, confounders were considered, but measuring bias might exist in self-reported data such as smoking status, alcohol consumption and daily fruit and vegetable intake. Furthermore, the quantity of alcohol consumption was not assessed. The results for alcohol consumption, daily fruit and vegetable intake and other comorbidities including diabetes, hypertension and cardiovascular disease are shown in the supplement (Table S2 and Table S3). Finally, the periodontal status in this study was diagnosed by dentists, not self-reported by the participants. Because our data are secondary data, we don't have information about the consistency of periodontal health assessment, in the form of agreement between dentists.

Within the same age group as our study, the mortality rates in Taipei and nationally in 2012 were 0.62 and 0.67% respectively. In this study the mortality rate in 2012 was 0.59%, which was slightly lower than the national mortality rate in the same period. The common reasons for non-participation among the elderly were unsuitable timing or location of the health examinations, lacking time, or feeling too sick to participate.[49,50] Because the health examinations in our study were performed by Taipei-contracted hospitals in each administrative division in Taipei, citizens had easy access. Nonparticipants in our study might be those with poorer health. Moreover, Taipei is a highly urbanized city, and has a mortality rate lower than the national mortality rate. In urbanized areas, age-standardized mortality rates were lower than those in suburban and rural areas.[51] Our estimates of hazard ratios might be slightly low.

Strengths of our study are the large sample, and the fact that these participants were followed across time. Over 30% of the sample were diagnosed as having periodontitis at the baseline. Regular health screening seems important due to the high prevalence of this disease. Health screening allows early identification of chronic non-communicable diseases and its risk factors, which easily happen in the elderly. Health screening can lead to the reduction of related complications as well as mortality.[52] Screening for health problems that can emerge in later life could reduce the burden of disease. Regular oral examination can detect the signs and symptoms of oral disease early, in particular dental caries and periodontal disease.[53] Lowering periodontitis prevalence may reduce the medical expenditures of health care systems.[14,54]