Public Health Implications of Chronic Periodontal Infections in Adults

Question and Answer Session: Day 2, April 9, 2003

Moderator: Gary Rozier, DDS, MPH

Question
Unknown – What definition of periodontal disease did you use?

Answer
Steve Offenbacher – The definition we used is an increase in pocket depth. You can use attachment loss, although there’s a much smaller fraction of mothers who experience attachment loss. From our standpoint, this is an infection and inflammatory process that is associated with an increased risk. One problem we have is that we are linking our diagnosis of a clinical condition based on an assumption that we’re lumping them into gingivitis or periodontitis, and these are driven by treatment codes and therapies designed to reverse or treat those conditions. If you look at the data, they do not sort out out cleanly along traditional definitions in terms of things that are most associated with the temporality of the systemic condition you’re trying to model. We have two papers coming out that address rethinking and clustering clinical signs. I don’t think we can categorically say that periodontal disease has to be attachment loss. It’s an infection and inflammation and a group of clinical signs that we can relate to the systemic exposure. Using clinical parameters to define it in a systemic exposure is a lot different than using clinical signs to describe it as a condition that relates to the longevity of the two. For example, bleeding on probing is an important marker of this wound area. We find that it relates very closely to temporal shifts in blood levels of inflammatory markers or things that require decades to change, e.g., IMT. The fact that this is gingivitis and not periodontitis does not bother me at all. Call it what you like. It’s a clinical sign of a worsening condition of periodontal health, as associated with changes in the flora, inflammation, and swelling of the gums. I don’t care if you lose a millimeter or not. It’s probably irrelevant.

Answer
Unknown – Dr. Jeffcoat, regarding your intervention study on periodontal disease and preterm birth, that is very nicely done, my question is on the characteristics of the women in the three treatment groups and the reference group. Because the women in the reference group are not passed through the randomization process, I wonder if the characteristics of these three groups are comparable in regard to the risk factor of preterm delivery.

Marjorie Jeffcoat – Yes, they were drawn from the same population…and the characteristics would come out the same if you looked at the percentage of smokers, those who drink, the amount of periodontitis they have, prior preterm births, etc. How did we get that group? That group occurred because we had so many volunteers to be in this trial (which says something, by the way, about access to care) that the Institutional Review Board (IRB) gave us approval to do a pilot study early on. They said, do the pilot, show it’s safe, then go ahead and do a bigger study, which made sense. So we took every third patient in the reference group who volunteered because there wasn’t enough space for them in the clinical trial. Then we followed up with phone calls to ask if they had had treatment and all that. But they aren’t blinded in the same sense at all. That is, they didn’t come in and we didn’t blindfold them and pretend to clean their teeth. So it gives you some sense of how much prematurity there should have been in that population if you didn’t affect it at all.

Kaumudi Joshipura – I have a comment for George Taylor. The nutrition part isn’t as important for the pregnancy outcome [since] nutrition isn’t as related or likely to be a confounder, but diabetes is probably likely to be much more important. It’s been shown that whole grain fiber is protective for periodontal disease and also for diabetes, and there could be several other common risk factors also. In the context of diabetes, nutrition is definitely important to evaluate. The other general comment—randomized trials are great and will give us a lot of answers quickly in the pregnancy outcomes area, and maybe not so quickly in the diabetes area. But, as public health practitioners, we should not lose our focus on preventing the occurrence of periodontal disease in the first place. With the treatment, you can reduce the impact of periodontal disease on the pregnancy outcomes, but there’s still the difference between people who have periodontal disease and those who don’t. That has to be emphasized. The cohort studies are equally important, and we also need to look at the mechanistic aspects.

Gary Rozier – I was struck by some of the data presented this morning. My understanding is that Latinos have a higher rate of periodontal disease than whites. On the other hand, their rate of adverse health outcomes is essentially equivalent to that of whites. A major goal of public health in this country is to reduce or eliminate disparities along racial or ethnic lines. Have any of the studies you’ve reviewed on birth outcomes and periodontal disease addressed this?

George Taylor – My understanding is that Latinos have a lower rate of prematurity that is very comparable to that of Caucasians, but after they move to the United States their risk increases. I believe there is a study investigating that phenomenon, just like the risk for heart disease is increasing. There is probably some relationship. I don’t believe those connections are really understood. We assume that when people move to different places, there are dietary and cultural influences. But in terms of the oral cavity, there are important microbial differences that have been found to be related when people have done whole biofilm analyses. They are finding geographic differences among races. For example, whites in Sweden are different from whites in Minnesota in terms of the oral flora that’s associated with their periodontal condition. We are living as a consequence of our environment, and these organisms are passed and have no other habitat but the oral cavity. As a consequence, we expect to see regional and ethnic and cultural differences in rates.

Question
Eugenio Beltrán – The issue of preterm labor, low birth-weight, appears to have very specific population demographics; mostly young women or those in their teens, which has different social aspects. None of the speakers has addressed this issue in their assessment of the risk. Maybe we have some background that makes these people more susceptible to periodontal disease and early outcome and pregnancy that we are not capturing.

Answer
M. Jeffcoat – I can speak to our intervention study. Twenty-five percent of our group was below age 20 and many were below age 18. We had IRB approval to capture people down to age 14. The mean age turned out to be within 1 year across the groups. We didn’t have anyone over 35. The group was 80% African American and 20% Caucasian, balanced across the groups. We only had three Hispanics of any definition in the trial because Birmingham didn’t have many Hispanics then. But it’s changing. Whatever we knew about, we tried to balance for.

Question
E. Beltrán – In your design of the observational study, do you keep in mind that some of these clinical situations that you can obtain by volunteering or going through the records, that these women have a specific social structure and characteristics that may make them more at risk of having behaviors associated with a particular outcome? I wonder if there are other real causes of this situation and not just periodontal disease.

Answer
M. Jeffcoat – It can be another process that you don’t know about. But in the adjusting the known comorbid factors were either adjusted for, or in the intervention trials, those patients weren’t allowed in the study. All you can ever do is design a study based on today’s knowledge.

Susan Reed – A paper by Pierre Buekens expresses his disillusionment over the state of research, in what we’ve learned, and what we don’t know in regard to preterm birth. The positive part is that research needs to be done about the mechanism, the biology of birth, which is a huge area. There’s so much to understand. The second area has to do with the social factors that, again, we don’t know about, because we’re not measuring them. They are equally not well understood.

M. Jeffcoat – We have a 2 hour psychosocial behavioral questionnaire that went along with the intervention study, but that database isn’t yet locked because they’re still making sure the entries are correct. I can’t analyze that before it’s locked.

S. Offenbacher – In our particular study, we did try to enroll all pregnant women so there was no selecting out of sub-populations. I agree with what Marjorie [Jeffcoat] said. We try to look at other potential causes of comorbidity, and there’s probably an underlying relationship of some risk in that the same patients bear the burden of disease. There also has to be some biological reason for these associations. In other words, why does taking someone’s money away make them have more disease? Being poor by itself is not a good reason to have more disease. Perhaps there’s physiological stress, a change in the immune system, and other exposures to other infectious agents. So, in this process the socioeconomic and other parameters only give clues about where these things are going, and there are still not biological explanations for these relationships.

If you think about periodontal disease, the data suggest that the level of plaque control does not predict who has severe disease in the population. The prevalence of severe periodontal disease in rural China where they do not brush their teeth or have dentists, is about the same as in Erie, Pennsylvania. We all have the same oral pathogens and are exposed to the same biofilm, yet certain people in the adult population develop severe periodontal disease. We believe it is due to the fact that these patients have inflammatory, immune characteristics that enable these organisms to take hold. You have a lot of inflammation in response to the flora, and therefore more disease. You have an inflammatory phenotype that would also put you at greater risk for other conditions, particularly those that have inflammation as part of the chronicity of their disease for pathogenesis, such as heart disease and complications from diabetes. So we have suggested in the past that periodontal disease is a way of sorting out people who have an inflammatory phenotype in the population because the mouth is a mirror to systemic health in this regard. However, that does not negate the importance of the oral infection as a burden in individuals who have this predisposition for advanced inflammation. If anything, it helps target those who are at greater risk.

So, yes, there is probably an underlying characteristic that places people at common risk for many inflammatory conditions. In 10 years, there may be a significant profile that defines these at their genetic basis. There may be differences in various populations, so there’s probably a large genetic component to this. But that doesn’t mean the infection is not an important part of the process or that it is strictly due to those differences that are clustered along demographic or socioeconomic borders.

M. Jeffcoat – In finding red gums in somebody with plaque who’s pregnant, is a diagnostic marker solely for the fact that you can turn out more cytokines, for example? Therapy shouldn’t matter to the ultimate outcome, such as preterm birth. But I had trouble coming up with a mechanism for how treating that redness gives you fewer preterm babies without somehow going through the treatment being in the causal pathway. I get it all the way up through the longitudinal studies. But once you start to intervene, then you’ve got something else.

E. Beltrán – Interventions to prevent disease or reduce the morbidity of certain diseases don’t have to be causal. You can intervene in one of the factors, i.e., you can find that periodontal diseases have a very strong association with your outcome of interest and you can say, “Let’s go do some intervention and see how it works.” But that doesn’t mean it’s causally affected. It’s just that when you are doing the intervention, you are somehow affecting the causal pathways and preventing the disease.

Summary Remarks: The CDC Perspective – William R. Maas, DDS, MPH, Director, Division of Oral Health, Centers for Disease Control and Prevention

Back to Chronic Periodontal Infections Conference

Historical Document
Page last reviewed: February 2, 2005
Content source: Division of Oral Health, National Center for Chronic Disease Prevention and Health Promotion