This was a fun journey.
The first couple of links didn't have any references, but the third one did. Not only did it propose a mechanism, but it ascribed it to "Dr. Perls" and "Dr. Roizen".
Picking the second doctor (with the biggest claim) at random. His name is Michael F. Roizen, MD. of the Cleveland Clinic. His claims are somewhat controversial - popular, but attacked by academics.
He wrote a popular book RealAge. The relevant section was quoted in more detail by this dentist:
"These studies show that the presence of periodontal diseases, diseases most common in people with tooth loss, actually affects longevity. The best of these studies done at Emory University with the Centers for Disease Control, indicated that people with gingivitis and periodontitis have a mortality rate that is 23 percent to 46 percent higher than those who don't... why? They are linked to increased rates of cardiovascular disease and stroke, as well as to an increase in mortality from other causes, such as infections."
So, I wandered over to the Center for Disease Control's Oral Health site, and looked for some references.
This seemed to be the most direct reference they provided:
- James D. Beck,Paul Eke, Gerardo Heiss, Phoebus Madianos, David Couper, Dongming Lin, Kevin Moss, John Elter, Steven Offenbacher, Periodontal Disease and Coronary Heart Disease: A Reappraisal of the Exposure, Circulation. 2005; 112: 19-24, doi: 10.1161/CIRCULATIONAHA.104.511998
which referenced an earlier paper by a similar team:
The 1998 paper makes some very tentative associations between periodontitus and coronary heart disease.
We conclude that the available evidence does allow an interpretation of periodontitis being a risk factor for atherosclerosis/CHD. This conclusion, however. is made with some qualifications. While the associations found across a wide variety of subjects are remarkably consistent, for the most part they are represented by incidence odds ratios around 2.0. While this level of association would result in oral conditions contributing to a large number of CHD cases, it is possible that associations of this magnitude are due to bias in the study designs. In addition, some studies report that periodontitis is associated with all-cause mortality and low birth weight infants. These multiple associations detract from the credibility of periodontitis as a risk factor, as specificity of association is more often related to causality. However, all-cause mortality may largely be driven by mortality from cardiovascular events: and some exposures, such as smoking. are indeed risk factors for multiple conditions. On the other hand, current findings regarding the associations between oral conditions and atherosclerosis/CHD imply that the criteria for causality may be met in the not-too-distant future.
The 2005 paper shows they are still working establishing this association (e.g. by continuing to try to remove cofounders, such as smoking, and work out the best measures of periodontal disease) and reviews some of the criticism it has received.
Case-control, cross-sectional, and longitudinal studies have found that periodontitis is associated with coronary heart disease (CHD) and cerebrovascular disease, even after adjustment for a variety of potential confounders of these associations. However, other studies have found either nonsignificant positive trends or no association after adjustment for variables considered to be confounders. Concerns about the validity of the periodontitis–cardiovascular disease associations have been expressed.
They go on to defend their hypothesis, but admit that standard clinical measurements of periodontal disease show no associations. (Instead, they recommend measuring antibody responses.) Again, their claims are heavily qualified.
There has been some interesting research tentatively suggesting that periodontal disease is closely correlated to coronary heart disease, with a mechanism proposed. That research has been criticised and (as of 2005), the science hadn't reached a consensus.
Roizen published a popular health book in 2000 (which spawned a series of books and a web-site business) that claimed periodontal disease could cause an elevated risk of coronary heart disease, without considering the confounding factors. He made the leap that flossing would therefore reduce the risk. His book was promoted on Oprah and gained widespread attention from dental web-sites, but it was widely criticized by scientists as being exaggerated.