Poor oral health and coronary artery disease: Association or causality?

Despite a trend of decreased annual death rates from coronary artery disease (CAD) in Europe, the prevalence of cardiovascular disease, and especially CAD, is still high in western, but also in developing countries.¹ Risk factors have been well defined including genetic predisposition, hyperlipidaemia, hypertension, diabetes mellitus, physical inactivity and smoking. Furthermore, poor oral health and especially periodontal disease have been repeatedly associated with CAD in observational studies; however, a causal involvement of periodontal disease for the development and progression of atherosclerosis has not yet been unequivocally proved.² All association analyses are confounded by the fact that several CAD risk factors also contribute to the risk of periodontal disease, especially cigarette smoking and diabetes mellitus. Thus the controversy is ongoing as to what extent pushing oral hygiene and treatment for periodontal disease is able to prevent cardiovascular events.

An important link between poor oral health and atherosclerosis may be vascular inflammation, either as a direct effect of oral pathogens in the vascular wall or by stimulating more globally an inflammatory environment. In fact, experimental gingivitis has induced systemic inflammatory markers in young healthy individuals.³ In the response to pathogens and/or inflammation, the endothelium is considered to be the prime target.⁴ Endothelial dysfunction and subsequent thrombotic events leading especially to obstruction of the small vasa vasorum may translate into functional impairment and structural damage in the vessel walls of larger arteries.⁵ Although oral pathogens have also been demonstrated in specimens of atherosclerotic plaques, the detection of viable microorganisms has been difficult (for review see Lockhart et al.).² In addition, trials of long-term antibiotic treatments in CAD patients, for example, in the WIZARD study, did not show a reduction in clinical events.^4,6 However, the CANTOS trial demonstrating a significant reduction of major cardiovascular events in patients after myocardial infarction by targeting inflammation using an interleukin antibody renewed the interest in anti-inflammatory strategies to prevent the development and progression of CAD.⁷

To shed more light on the still unanswered question of a potentially causal connection between CAD and oral health, the report by Batty and colleagues⁸ in this issue of the European Journal of Preventive Cardiology is an important step forward. These investigators analysed the association between poor oral health and subsequent CAD morbidity and mortality in almost one million people included in the large Korean Cancer Prevention study, as a part of which all participants had an oral examination with a count of lost teeth at baseline.⁸ Follow-up was 20 years, with CAD diagnosed using national mortality and hospitalisation registers, and led to the analysis of an impressive number of more than 60,000 events. The main finding of the study was a moderate but highly significant association of tooth loss (seven or more teeth) with CAD events that was similar in men and women, thus supporting previous smaller analyses. The distribution of several established risk factors of CAD varied according to tooth loss, that is, some factors were more frequent in patients without tooth loss (physical inactivity, family history of cardiovascular disease), whereas others were less frequent in this group (smoking, hypertension, elevated cholesterol, diabetes). Next, the authors analysed the effect of the most important confounding factor, that is, smoking. Interestingly, when restricting their analysis to never-smokers, the association persisted in women, but was no longer apparent in the male population. Although this gender difference appears to be clear at first glance, the lack of confounding effects of smoking in women has to be interpreted with caution as only a very small minority of Korean women are smokers.

What to conclude from this analysis? First, several limitations of the study have to be noted: the main endpoint CAD was retrospectively determined from health insurance claims and death certificates with their notorious uncertainties. The description of oral health was based only on the number of lost teeth, which is only one indicator of oral health, albeit clearly related to periodontal disease. However, this minor criticism does not detract from the value of the main result of this analysis in more than one million people, that is, that confounding by smoking largely explains the link between poor oral health and CAD. The strengths of the current analysis are the huge number of patients, the long follow-up and the high rate of almost 65,000 CAD events. Furthermore, the transferability of the results to other populations outside Korea appears to be likely as associations known to be causal in western populations, such as blood glucose levels in relation to CAD and smoking in relation to several cancers, have been replicated in the dataset of the Korean Cancer Prevention study.^9,10

What advice can the busy doctor give for the primary and secondary prevention of CAD? 1. Although the effect on CAD and cardiovascular events is not unequivocally proved, proper oral hygiene and professional treatment of periodontal disease is at least beneficial to prevent tooth loss. In addition, a surrogate marker of inflammation such as C-reactive protein as well as endothelial function known to be related to CAD are significantly improved by periodontal intervention. 2. Abstaining from smoking is the most important means for the prevention of both periodontal disease and CAD, thus major efforts must concentrate on smoking cessation. 3. Physical activity is related to improved endothelial function, reduced inflammation and a lower risk of CAD. Interestingly, moderate and severe periodontitis are independently associated with low cardiorespiratory fitness.¹¹ In addition, enhanced physical activity is ‘rejuvenating’ and is associated with reduced biological age as measured by telomere length.¹²

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: JB and DB received support for original work regarding physical activity and ageing from the Deutsche Forschungsgemeinschaft (Cluster of Excellence REBIRTH; REBIRTH active). DB received a grant from the Deutsche Stiftung für Herzforschung (Project nr. F/27/12).

References