Home Product Information Clinical Studies Store - Purchase a Purebrush and Accessories Testimonials

Cross-sectional studies performed with convenience samples, such as this study, cannot be generalized to the entire population. However, they can identify as risk indicators those parameters that are significantly associated with the condition being investigated (in this case, CHD). Whether these risk indicators prove to be risk factors (that is, causally related) remains to be demonstrated in prospective studies performed with a more representative population.

The results of this cross-sectional study confirmed those of previous studies conducted with younger subjects that showed a significant association between dental disease and CHD.1-6 We found that the subjects with one to 14 teeth were most likely to have CHD, and observed this association in subjects in both independent living and dependent living circumstances.

Chronic infection hypothesis. The importance of having only a few teeth in these subjects seems to be related to the greater periodontal morbidity of their teeth�that is, a higher percentage of teeth with attachment loss�to poorer oral hygiene as assessed by higher plaque scores, and to less-frequent visits to the dentist (Table 3). In the multivariate analysis, a high BANA test score was positively associated with CHD in both the all-subjects dentate model (Table 4) and the independent living dentate model (Table 5).

The BANA test detects the presence of an enzyme(s) that hydrolyzes the synthetic trypsinlike substrate benzoyl-DL-arginine-naphthylamide, or BANA, in plaque samples.16 Of more than 60 plaque bacterial species that have been tested, only P. gingivalis, Treponema denticola and Bacteroides forsythus always exhibit strong BANA activity, although some species, like the Capnocytophaga species, occasionally exhibit weak BANA activity.16 The three strongly BANA-positive species are invariably elevated in plaque samples removed from teeth with periodontal pathology.17-20 Thus, a higher plaque BANA score would indirectly indicate that these periodontopathic species are elevated on the tooth surfaces of subjects with CHD. This implies some degree of periodontal pathology, which was documented by the significantly higher PBS in the multivariate model of independent living dentate subjects (Table 5).

These BANA-positive species are gram-negative anaerobes, so that their elevation in the dental plaque would support the various hypotheses linking chronic bacterial infection to CHD via effects mediated by endotoxins or lipopolysaccharides, or LPS.21-23 The confrontation of the varied host-defense mechanisms with these LPS-containing invaders may lead to the supplementation of serum cytokine levels with inflammatory mediators derived from this confrontation,24 which could contribute to the chronic inflammatory process that leads to increased levels of C-reactive proteins in the serum.8 LPS has long been known to promote atherosclerosis and thrombus formation.25

An additional factor in this process could be an exaggerated host response to LPS, mediated by the presence of hyperresponsive monocytic cells.6 Certain patients with early-onset perio-dontitis, refractory periodontitis or insulin-dependent diabetes have peripheral blood monocytes that secrete threefold to 10-fold greater amounts of PGE2, IL-1b and TNFa when exposed to LPS in vitro. These people would react to the penetration of gram-negative bacteria, such as the BANA-positive species, from the subgingival plaque into the periodontium by the overproduction of cytokines that could leak from the gingival tissue into the general circulation and have an effect on distant organs.

Other studies have proposed a chronic infection hypothesis involving dental disease as a possible risk factor for the development of a cardiovascular event.21,22 In their prospective study, Mattila and colleagues3 were able only to associate dental infections as measured by the TDI and a history of myocardial infarction with a subsequent acute myocardial infarction. They then showed that there are elevated levels of von Willebrand factor antigen in subjects with an elevated TDI,26 which could reflect increased endothelial cell damage induced by the LPS derived from the gram-negative dental plaque flora. Grau and colleagues7 have shown that the TDI may be associated with an increased risk of cerebral vascular ischemia.

Missing teeth. The number of missing teeth was independently associated with CHD in the two epidemiologic surveys that randomly selected their study populations,4,5 but these studies did not distinguish between edentulous subjects and dentate subjects with many missing teeth. Our results indicate that in these elderly subjects, it is the presence of only a few teeth that is associated with CHD. When we divided our subjects into groups with one to 14 teeth and 15 to 28 teeth, the presence of one to 14 teeth was significantly associated with CHD in both the all-subjects and dentate subjects models (Table 4).

An indirect confirmation of the importance of having 15 to 28 teeth in protecting against CHD was the relationship between the salivary levels of S. sanguis and CHD status. The significant negative association seen in all models (Table 4) can be explained by the fact that this organism is associated with teeth, and is among the best microbial colonizers on the tooth surface.27

This organism is dominant in early plaque formation, so that people with good oral hygiene would have higher proportions of S. sanguis in the mouth. The more teeth in the mouth, the more likely that the salivary levels of S. sanguis would be higher.
S. sanguis has been studied for its role in endocarditis,28,29 and a platelet-aggregation-associated protein has been identified as a virulence factor because it interacts with platelets, promoting coagulation, or it may behave as a heat-shock protein, which, by causing an autoimmune response, could initiate the early atherosclerotic lesion.30 Our findings do not support a role for S. sanguis in CHD in these older subjects, but it is possible that such interactions could have occurred earlier in their lives.

Evidence in the literature suggests that edentulous people with and without dentures and dentate people with missing teeth change their eating habits after they lose their teeth.31,32 They may avoid certain nutritious foods because of difficulty in chewing, and select high-calorie, high-fat foods whose consumption is recognized as a risk factor for cardiovascular disease.33 But an additional feature of this dietary change that could be particularly relevant in those dentate subjects with one to 14 teeth is that dietary-induced elevation of serum low-density lipoprotein levels has been shown to increase monocytic responses to LPS.34 These subjects would have both the dietary-induced sensitization of the monocytes and the plaque-laden teeth and gingivitis that could provide the LPS challenge to these cells.

The advanced age of our subjects might explain the absence of an association between edentulism and CHD in the logistic regression models, since people most susceptible to the linkage between edentulism and CHD may already be dead. This would be consistent with the National Health and Nutrition Examination Survey, or NHANES, data that showed that in males younger than 50 years of age, edentulism was 2.6 times more likely to be associated with death resulting from any cause.4

However, this explanation, as well as the chronic infection hypothesis involving dental infections, would not seem to explain why 53 percent of our edentulous subjects had CHD, unless these patients experienced their first heart attack while they had teeth. It is possible that by having had their teeth extracted, these people coincidentally had the tooth-related microbial challenge removed that had been a major contributor to the events leading to their heart attack. Edentulism in this sense would be protective, although in the younger subjects in the NHANES study, edentulism might have been a marker for rampant dental disease that could have predisposed them to CHD by the mechanisms described above.

Other significant findings. The possible protective role of daily alcohol consumption noted in the all-subjects model in this study has been observed by others.35 The greater use of all types of medications was significantly associated with CHD in both the all-subjects and dentate subjects models. This was expected, as our subjects were taking many types of medications associated with old age and a diagnosis of cardiovascular and other diseases.36

It is not clear how a complaint of xerostomia could be such a powerful risk indicator for CHD. Xerostomia is a common complaint of older people,37 and it has been associated with poor oral hygiene, the inability to chew and involuntary weight loss among both institutionalized38,39 and independent older adults.38 We have found that people with complaints of xero-stomia avoid crunchy (carrots), dry (bread) and sticky (peanut butter) foods, but not crumbly (cake) or chewy (red meat) foods.40 This choice of foods could lead to the selection of the high-calorie, high-fat foods that are associated with obesity and CHD.33

The use of a convenience sample of older subjects for our analysis may explain why recognized risk factors, such as serum cholesterol levels, BMI, smoking status and diabetes, were not more prominent in the modeling procedure. The use of medications to control blood pressure or cholesterol levels was not included in our analysis; as these subjects were actively receiving treatment for CHD, this could explain why these risk factors were not associated with CHD.

In other studies performed with convenience samples, the dental variables were significantly associated with CHD, whereas the classical risk factors such as cholesterol levels, alcohol consumption, BMI and hypertension were not.1-3,7 Thus, our investigation, like others performed with convenience samples, cannot be generalized to the entire population, and because of the cross-sectional nature of our study, the oral health variables can at best be considered risk indicators. Despite these limitations, for this group of older veterans, our findings suggest that the oral health parameters may play an important role in the occurrence of CHD. We are following up these subjects to determine whether any of these dental parameters can be shown to be risk factors.




Home | Information | Clinical Studies | Press Releases | Store | Testimonials

123 Primrose Road - Burlingame, CA 94010
Sales: 800-439-2497 Lab: 650-579-1352
 Fax: 650-579-1351 Email: