-Hello, everyone,
and thank you all for joining us for our CDC Dental Public Health
lecture series. My name is Gina Thornton-Evans, and I'm the director of the CDC Dental Public Health
Residency Program and also a team lead
for the Surveillance, Investigations,
and Research team within the Division
of Oral Health at CDC. Today, it is my pleasure to
introduce Dr. George Taylor. Dr. Taylor is a professor
in the Department of Preventative
and Restorative Dental Sciences in the Division
of Oral Epidemiology and Dental Public Health. He is also the Associate Dean
for diversity and inclusion at the University
of California, San Francisco School
of Dentistry. He is a board
certified specialist in dental public health. He began his career as a dentist in the U.S. Air Force
for four years, and for the past 40 years, he's been in
an academic setting, teaching, conducting research,
and providing patient care.
In his role as an associate dean
for diversity and inclusion, he works to strengthen
the School of Dentistry as a diverse
and inclusive climate that empowers students,
faculty and staff to maximize their capacity
in learning, working, serving,
and growing together. Dr. Taylor's major research
focus is on relationships between oral
and systemic health, particularly periodontal
infection and diabetes outcomes. He is the past president
of the American Board of Dental Public Health, and he's also served on the
Council of Scientific Affairs of the American
Dental Association and the NIDCR Board
of Scientific Counselors. Additionally, he actively
mentors students, residents, and other faculty
in their careers with respect to research. And now I will turn it over
to Dr. Taylor. -Thank you,
Dr.
Thornton-Evans. Certainly an honor
to be here today to share some of the — some of
my learnings with the audience. I appreciate your introduction
and I'm going to go ahead and thank you for all
who are attending. And welcome today as well. I'm going to start sharing
my slides and I'll share my screen
and I'll go from there. May I just confirm that you
can see my — my first slide? -Yes.
-Okay, thank you. Thank you. Well, today is my — really my pleasure to talk about
diabetes and oral health and diverse
research perspectives. I'm going to, uh… We're going to have a journey
where I describe study designs that are appropriate
for addressing different kinds of questions
that will be intertwined with talking about
the state of evidence with respect to
many different associations between diabetes
and oral health, going in — actually, going in
both directions as well.
So I'm providing this slide
for members of our audience who are not dentists to get a visual picture
of periodontal disease. And in this case,
we're going to — reason I show
periodontal disease first is because it is one of the
major oral health conditions that contributes to a
bi-directional relationship between diabetes
and oral health. And just as a review, and perhaps some
additional information, you see here in this slide two forms
of periodontal disease: the top side is gingivitis. This is where
the gingiva is inflamed, and yet there's not any loss
of the supporting structures of the periodontal ligaments
or the alveolar bone that supports
all of the teeth.
You can see the gingivitis and then you can see
a very severe periodontitis. This happens to be a person
who has diabetes, and you can see
just about every single aspect of oral health as —
or, poor oral health as it pertains to the teeth
and the gums. You see gingival recession, you see loss of —
loss of support where the gingiva
would normally be here, there's recession, you see very poor oral hygiene.
You see very —
edematous as well as — as well as tissue that is
very inflamed. Just to get a contrast, now here's a healthy
periodontal tissue. So this is —
this is a patient who does not have gingivitis
or periodontitis. Here again,
very severe periodontitis. And in between
you see gingivitis, swollen — swollen tissue, but not severe —
not severe recession. Here we see a little recession
in this area. So this person has
very localized periodontitis, mostly gingivitis, and in this case, generalized
severe periodontitis. So we're going to see
some other slides as I'm going through, but first of all, I'd like to start talking about
evidence from the cross-sectional
point of view.
And of course, we know that
cross-sectional studies are where measurements
are all taken at the same point in time, and in our situation
the controls would be people without diabetes, other people without
periodontitis, cases would be those with
periodontitis or those with diabetes. And I mentioned both of those because we are thinking about
a bi-directional relationship. And then we analyze those
studies from the point of view of amongst the people of the control group, if you will,
those who are exposed or not exposed to the exposure, which might be, in this case,
periodontitis or tooth loss.
It might be diabetes. It might be those with poor
hyperglycemic control. It might be those
with gestational diabetes. Those are some of the things
we're going to talk about. So I'm going to first talk
a little bit about the evidence for as it pertains to periodontal health as well as diabetes
in different — in different aspects
of diabetes status. So this is an older slide. And yet the proportions
and the prevalence of moderate and severe
periodontitis by age remain about the same,
and you can see a gradient: as age increases, uh… as age increases,
so does the — so does prevalence of moderate
or severe periodontitis. Moderate periodontitis here, severe periodontitis here; lower proportion
of the population has severe periodontitis; a much greater proportion
of the population have — has —
the moderate periodontitis — except when we get to — And then you can see
a summary here.
About eight-point —
eight percent have severe periodontitis. About 30 percent
have moderate. To the right of the slide,
these are — these are the definitions for the American Association
of Periodontology, Centers for Disease Control, the method for assessing
periodontal disease has been changed,
the classification has been changed just a little bit. And yet the proportion —
as I said, these prevalences
are about the same. So then, first, we're going to
look at diabetes and prediabetes and their adverse effects
on periodontal health. Here is a striking slide,
I believe — as you can see, we have the categories of those who have smoking status of
never smoker, former smoker,
or current smoker. Now the reason that we put
our participants in the study — this is an enhanced study — the reason is because
tobacco use, particularly smoking, is one of the most
prominent contributors to poorer periodontal health. And yet,
as you can see in each category, from never smoker to former
smoker to current smoker, we can also see a gradient
for diabetes status.
This is normal glycemia, prediabetes,
and then diabetes. In each category,
you see — we see the same gradient. So that's the impact
that diabetes has on being associated with
poorer periodontal health. So then we would go —
look at the next slide — and so those were adults. Now diabetes has adverse effects
on periodontal health in children
with Type 1 diabetes. Gonna talk a little bit
about that. People — adults with Type 1
or Type 2 diabetes — a shared study
with that respect — and also gestational diabetes. So then first of all, we're going to talk
about a case control study conducted at
Columbia University, and I thought it would be
helpful to share this image, where in a case control study,
cases identified — in this case it would be those
with periodontal disease, and control would be those
without periodontal disease — and we could say
good oral health.
And then from there we looked to
see who was expo– who was exposed
amongst the controls, who was not exposed. And then we do the same thing
with the cases. So the point is that
we first identify cases and then look backwards to determine
what their exposure status was. That's the way that we design
a case control study and then we performed
the analysis that way. So then this slide shows a study by E. Lalla
and colleagues at Columbia, and it is about
the percent of children with periodontal
attachment loss greater than equal to
2 millimeters, and greater than or equal to
two teeth. As you can see, these are the controls
and these are the cases. What's most striking
in this slide, and here — and most — the two age groups, six to 11
and then 12 to 18 years old. What's most striking is
the difference in the proportion of people — of children —
children and adolescents — who had no teeth with
a periodontal attachment loss — in the case of the controls
for the six to 11 year olds, you can see
the dramatic difference, a much smaller proportion who had attachment loss or had no teeth that had
attachment loss in the cases.
So that is one of
the important findings that the Columbia group have made and have probably
several papers on that. So then we're seeing that
there's an association between diabetes — Type 1 diabetes
and periodontal health. Then if we look
a statistical analysis, we can see that among all ages
combined in the group Lalla was studying, the odds ratio for those having
poorer periodontal health, in this case, two or more teeth
with attachment loss, was five fold greater for those
with diabetes than those without diabetes. And also in that model,
12 to 18 year olds had a 4.8 greater odds ratio for attachment loss of
2 millimeters or greater when compared to the — Those odds were
4.8 times greater when compared to
the six to 11 year olds.
They also conducted
separate studies, and we saw — and they were able
to see that in the 11 year old —
in six to 11 year olds, the odds ratio was 3.44. All of these odds ratios
are statistically significant because there's no one
involved — if there was a value of one,
that meant that — that would mean that there
was no difference. In the 12 to 18 year olds,
you see 20 fold higher odds and a very large
confidence interval that doesn't include one. This is probably
due to the sample size and the way that people were — the prevalence of people filling
some of the different groups. But the point is that
in children and adolescents with Type 1 diabetes
in their studies, they did find an important and significant association between the — in the case
and controls for having poorer
periodontal health.
Next, I want to spend
a little bit of time talking about research,
and we'll — with respect to case con– with respect to prospective or
cohort studies. Cohort studies can also
be retrospective. In the cohort study, we first assign people determine
the person's exposure. In this example,
it uses smokers. So the comparison groups
are identified and the sampling is done
on their exposure status. In this case, there's a group who are exposed to
tobacco use, and there's also a comparison
group that is not exposed. They are followed over time or they are looked
at retrospectively, and then they compare
the outcomes. As you can see, a greater
proportion of the of the images, the stick figures,
have the outcome — whatever that outcome
might be — let's say it's poor
periodontal health — than those
who were nonsmokers.
And you see one
as opposed to three. So then again, the point is
that cohort studies allow us to follow
our participants over time or to look retrospectively. But the key and the distinction
between cohort studies and case control studies is that case control studies
start with the cases or sample on the case —
those who are cases and who are not cases to
select them into the study, whereas the cohort study of
samples on the exposure — who was exposed
and who was not exposure. Now we're going to take a look at some prospective
cohort studies. This radiograph is — actually the top figure was
at baseline. The lower radiographs
are at two years subsequent. This is two sets
of radiographs from the Gila River
Indian Community, and I want to respect
and acknowledge the members of the Gila River
Indian Community for participating in
this long term study of — that the NIDDK conducted
for examining and learning about periodontal — learning
about diabetes, particularly and in this case,
specifically Type 2 diabetes.
The late Dr. Bob Jenko was the principal investigator
for the oral health component, and I was very fortunate
to be able to join that team when I was working on
my doctorate in public health. And so I have had
quite a bit of experience with the Gila River
Indian Community's data. And this shows us —
the point for this slide is to examine just how much and how rapidly periodontal
tissue destruction can occur in people with
Type 2 diabetes. And then if they're more —
if they're poorly controlled, then that actually makes
the destruction more severe and also more rapid. Here's the panoramic radiograph, where you can see
all of the teeth at baseline. The key is to look at where
the alveolar bone support is. Here we can see
extensive resorption of bone typically in the bone level
was stopped right at just below the junction
between the crown and the root, in this case at baseline
there was already bone loss. And over here there was even —
even more bone loss. So what happened was, you can see, this is
the amount of bones supporting the tooth at baseline.
Then when we look two years
later, look at the difference. There is a great — There's a great destruction
of the alveolar bone. All of this lucency, all of this
dark gray area here represents bone loss. The bone loss also occurs
between the two roots. We call that the [indistinct]. You see that the bone
is full at baseline. Two years later, we see
just rapid destruction and loss of tissue. This is a manifestation. And as [indistinct] said
in 1993, considered periodontal disease
to be the six complication
of diabetes.
So with that in mind,
we want to look at what our analysis of the
incidence of alveolar bone loss after two years
in a cohort study was, and the three groups
that we compare are those without diabetes, those with diabetes
in better control, at that time,
the glycemic control, the level of Hemoglobin A1C for
a cut point was nine percent. Today, that cut point, in terms
of clinical management, is seven percent.
So that would not probably
have made much of a difference. And yet this was —
these were the values we were using at
the time of this study. It was a while ago,
as you can see — in 1998. And what we see is 41 percent
of people without diabetes demonstrated alveolar bone loss using those
panoramic radiographs that you saw
in the previous slide. And in comparison,
57 percent of those with a Hemoglobin A1C less than nine percent
and having diabetes, and of course, there's a dramatic increase
in those who had diabetes and were poorer controlled
as well.
So this was one of
the studies that we were able
to see that diabetes and glycemic control
are important factors in loss of supporting
alveolar bone. Now I want to show you
another study. This study was by Ryan Demmer and his collaborators
at the time — they were
at Columbia University. And here we have people
who were followed over time and — in a cohort study — and this is the amount of change
in attachment loss over time. So here are people
without diabetes. The interesting thing
about their study was that it included adults
with both Type 2 diabetes, as well as Type 1 diabetes. So what happened over time,
if we look at people who had poorly controlled
Type 2 diabetes, we see that their bone loss
was much greater than those who had
controlled diabetes. So here's the contrast. This is over the same
period of time of follow up in terms of bone — attachment loss,
not bone loss, it's the attachment loss,
and the attachment actually is composed of
the periodontal ligaments, that attach the alveolar bone
to the root surfaces, so that when that attachment
starts to recede, we call it attachment loss.
Then if we look at people
who had — adults with Type 1 diabetes, and this is one — there are several studies
in Scandinavia that include adults
with Type 1 diabetes, and yet here is the study
that shows the — sorry — here we have
uncontrolled people with Type 1 diabetes, poorly
controlled Type 1 diabetes. And and here's
what it was for… Oops. And this is in contrast
to those who had well-control Type 1 diabetes. So it's not just the diabetes
that makes a difference, it also is the degree of control
of the diabetes that has an impact over time on the destruction of
periodontal tissues. We're going to go
to the next slide. Now we're going to spend
just a few moments talking about
gestational diabetes. And we're back to
cross-sectional studies now, and this is a set of
cross-sectional studies that…
I just need to move… I just need to move, um… I have a bar with some menus
covering my title. I know what it is,
so I'm not going to use our time moving that. But anyway, these are
several studies that actually investigated the association
of gestational diabetes with the prevalence
of periodontal disease. And as you can see
consistently there — there's a higher proportion
of those who had — women who had
gestational diabetes having periodontal disease
prevalent than those who did not
have gestational diabetes. And the odds ratios are
all reasonable odds ratios. There's one that actually
showed a protective effect here in this case. And because the odds ratio
is a fraction. But the overall view was that
that there was a significant association. And that's all what we could
say is association at that point in time —
those points in time — where the odds ratios
are here, and here are the 95 percent
confidence intervals, or they reported a P value,
which would mean that the 95 percent
confidence interval did not include one.
So these are — these were
early studies to — to — to suggest and show
in their study populations that periodontitis
and gestational diabetes were associated. And from that —
from those findings there evolved a series
of studies that were systematic reviews
and meta-analyses. To identify the study, look at the results of
of — of several studies, as you can see
in this diagram; combine those results
and do quantitative analysis on those — on the —
on the group of studies, as opposed to the slide
previously where I showed you individual studies. So this slide
is a recent meta– It was a meta-analysis by Kumar
and colleagues in 2018. The most important points
of this slide are, here on the list of the —
of the study. Here's a list the over
to the left of the slide, are the authors,
the year of the studies, and also the number of cases and controls and —
control members of the group. Over here, what's important
about these — these figures as the odds ratio, so we're looking to see
if the odds ratio — if there was a positive
odds ratio greater than one and that the 95 percent
confidence interval did not include one.
There are varieties there. The most important part
of this slide is this plot. This is the forest plot,
and the forest plot gives you the point estimate,
which would be the odds ratio, and then shows you the
95 percent confidence interval for each of the studies. This is a mixture
of case control — I'm sorry,
of cohort studies, as well as —
as well as longitudinal studies. Now the — the absolute
most important thing about this in summarizing it is that this
is the quantification here. And where this diamond
exists, then determines whether or not
the sum of all of the findings is — results in a statistically
significant association between gestational diabetes
and periodontitis. Where — and this line
is the null line, which would say there's no —
there's no significance.
As you can see, the diamond
does not touch the line. Therefore, there's a —
this forest plot tells us high statistical
significance. P is equal to .005, that there is an association
between gestational diabetes and the prevalence
of periodontitis or periodontal disease, depending on how the authors — depending on how the authors
defined it. So then what — Now I'm going to look at —
share some information about diabetes, periodontitis– prediabetes and tooth loss. Here, where this is two kinds
of tooth loss. One is one type of tooth loss
we call partial tooth loss. The other type of tooth loss we
call complete edentulism. So in —
in Nhanes's data, we've conducted an analysis
to determine whether or not there are
any significant differences between people without diabetes, those with prediabetes,
or those with diabetes and the mean number
of missing teeth.
And in this case, again,
we see that gradient. We would expect to see
a gradient by age with the the number of missing
teeth increasing as age increases,
as we see here. And then what —
what we also see is a gradient whereby
diabetes status or non-diabetes status has
a gradient — has that same kind of gradient. First overall of the
population in the — who were participating
in this study or who were allies
in this study. We see significant differences
from compared to those who were did not have diabetes
or normal glycemic status. For everybody summarized. And we see here diabetes,
here's prediabetes, and diabetes significantly — had significantly greater mean
missing teeth.
And then in those 65+,
we see that for diabetes, those with diabetes had a significant greater
number of mean missing teeth. So now we get
that missing missing teeth. Why would we want
to study missing teeth? Because people missing teeth, particularly if they have
diabetes, there could be
difficulty with chewing. There could be difficulty
with their diet selection. There are also, of course, social and other functional
impairments with people and teeth, and large numbers of teeth
start to miss. So it's important. Missing teeth, particularly
people with diabetes, is an important — another important
oral health dimension that we pay close attention to. So coming back to the part —
to the Pima Indians study, we look at those,
compared those with no diabetes to those who had
type 2 diabetes.
And again, in each age group, we see that
there is a difference in terms of the — and we see
a gradient by age again. That's constant. And we see that there are
much higher proportions in many of the age groups for those having
partial tooth loss, any partial tooth loss
in this case, between those who had diabetes
and those who did not. It levels off in the older
age groups in the Pima Indians. I have a trigger mouse, and when my hand gets
close to it, it reacts. So anyway, as you can see, the gradient is attenuated
quite a bit. And that's because
in the Pima Indians, there was considerable
tooth loss in both those with
and without diabetes in the older age groups. So here… …we're looking again at — this is risk for losing one or more teeth.
And again, this is
the Health in Pomerania study in Germany. And what we can see again are those individuals
with type 2 diabetes… …controlled
and type 2 diabetes — type 2 diabetes uncontrolled. We have type 1 diabetes
controlled and type 1 diabetes
uncontrolled. And what we see in each
of the comparisons is that those who were
poorer controlled with their diabetes —
and these are adults, again — had greater risk for losing —
for tooth loss. And we see it
in the type 2 diabetes. We see it
in the type 1 diabetes. And we also see it in those
who had incident diabetes. There's a difference between
those who were diabetes free. And also, I already spoke about the differences in the control. So diabetes controlled is an important characteristic
to keep in mind as we conduct analysis of the relationships between
oral health and diabetes. So then looking at
partial tooth loss, looking at completed
edentulism, again, we look at the prevalence
of edentulism by diabetes status and age
in U.S.
Adults. Again, we see that pattern. We see a pattern —
not as striking in some of the other slides. But overall,
there's a pattern whereby those who have diabetes have a greater prevalence
of edentulism than those who do not have —
who either have prediabetes or no diabetes at all. The place where we saw
the statistical significance was in the 45- to 64-year-old
group. And yet, there still
is this gradient of tooth loss by age. And then in this group, the 45 to 64 years,
the 65-plus-year-olds, we see not only
the gradient by age, but we see a tendency
for a gradient based on diabetes status. Interestingly,
those who had normal glycemia actually have more edentulism, higher proportion of edentulism,
than those who had prediabetes. We'll need to look further to be
able to explain that difference. Now we're going to change gears. Another diverse perspective
is actually periodontal infection. Turning things around,
we're looking at now periodontal disease,
periodontal infection, and its adverse effects
on glycemia in people who do not have
diabetes, as well as some diabetes
outcomes.
So then the inflammatory burden is what plays a major role in periodontal disease impact on diabetes
and glycemic control. So there's a —
we have a hypothesized model and also substantiated
by various laboratory tests that help us to understand
why periodontitis or periodontal disease,
which would include gingivitis and periodontitis,
have a pathway. And we'll look at
that pathway now. So then one of the things
is that periodontitis, as you can see on the right,
the inflamed tissue, the calculus leading to accumulation
of bacterial plaque. And that bacterial plaque
then produces products, and also the inflamed tissue,
as you see the red — and the red gingiva here, is an image of inflamed tissue
and you see the bone loss. This inflammatory state — and it's a chronic
inflammatory state — has several characteristics
that are similar to the chronic inflammation
of visceral obesity. So it produces
a pro-inflammatory state in which there is a chronic
overexpression of cytokines, and they would be
inflammatory cytokines. Those of importance
and common to obesity, as well as periodontitis
and the inflammation and an inflammatory response
associated with it are interleukin 1 beta, interleukin 1,
interleukin 6, Tumour Necrosis Factor alpha.
And those, as you can see —
those mediators can contribute
to insulin resistance. That is the inability
for the insulin to stimulate the cells to be able to absorb glucose
from the bloodstream. And also, in addition to contributing to
insulin resistance, those inflammatory mediators, again, disseminated
through circulation through the inflamed
periodontal tissues, also activate the liver to produce a set
of inflammatory reactions called the Acute Phase Response
for the liver. That would include
C-reactive protein, fibrinogen, plasminogen
activator inhibitor. And those also contribute
to insulin resistance. So you can see there are
a couple of pathways at work in different parts
of the body as periodontal inflammation disseminates
its bacterial products, as well as
the inflammatory mediators.
And insulin resistance
then contributes to impaired fasting glucose,
impaired glucose tolerance, which capture not only
diabetes diagnosis, but also prediabetes diagnosis. Also, insulin resistance
leads to diabetes and also poorer
glycemic control. And similarly, the acute phase
reactants from the liver contribute to those same
unfavorable outcomes that we see in the bottom right. Additionally,
the pro-inflammatory state actually creates
pancreatic beta cell damage and reduces
the amount of insulin, or eliminates the amount of
insulin actually being produced. So from that theoretical model, there's empirical evidence with respect to periodontitis
and insulin resistance. And I'm going to share
the results of a cross-sectional study first using enhanced data
from '99 to 2004. These were U.S. adults
who did not have diabetes.
And exposure was
periodontal disease, and it was the quartiles
of mean probing pocket depth, as well as using
the CDC-AAP definition for no, mild, moderate,
and severe. So the comparison was with
the first quartile of mean pocket depth,
or the control group was those who had no or mild
periodontal disease. The outcome was the HOMA
insulin resistance measure. HOMA stands for
homeostasis model assessment of insulin resistance. And the results of that study
showed that periodontitis was associated with the HOMA
insulin resistance measure, and the relative risk
that they estimated here was 1.24 times
higher association than those who did not have — those who were in
the lower quartile or had less periodontal disease. And looking at those
with severe association — severe periodontitis as defined
by the CDC-AAP definition, those who demonstrated
HOMA insulin resistance, their association,
or the relative risk, was a two-fold greater
association than those who had mild
or moderate — or no or mild periodontitis.
So this is a start
to empirical evidence. We can't make
a causal association because it's
cross-sectional study. And yet, these models
were adjusted for other covariates
and potential confounders. So that, again, adds strength
to the evidence that there is an association to the suggestion
that there's association. So then, I see my time is — I'm going to pause for a minute,
check in with Gina.
My time is running tight, and I want to make sure with
Gina that everything's okay. And I would like you to
let me know when I should stop, and we'll go with
question and answers. So….
-Okay. Okay, Dr. Taylor, we have — if you can,
maybe another five minutes. We do have a few questions
already. -Okay. All right. So another study investigated the association
of periodontal infection with impaired glucose tolerance. A glucose tolerance test,
of course, is one of the ways to identify those with diabetes or those with prediabetes. This, again, was another
cross-sectional study, NHANES 2009-2010. And these were U.S. adults
who were diabetes-free. The exposure was
periodontal disease, and the comparison group
were those in the 75th percentile
of probing depth or had no or mild
periodontal disease. The outcome, again, was
impaired glucose tolerance or impaired fasting glucose. And the results and conclusion
found a statistically significant
association between periodontal infection
and impaired glucose tolerance. The odds ratio here was 1.93 with a 95% confidence interval that did not include one.
So similarly, using the 75th
quartile as the exposed group, the probing pocket depth
was greater than that 75th quartile. There was also
a significant association, as you can see
in the odds ratio. All right, so this again — this study was adjusted for
other important characteristics. But then we'll move
to periodontal disease and dysglycemia development,
but not diabetes. So there have been
a couple of studies that have actually followed
people in cohort studies for 5 years or 10 years. And you see in this slide the
results of those studies were that hemoglobin A1c — the measure of glycemic control that hemoglobin A1c
provides for us was greater after five years in those people who had
experienced periodontal disease.
I'm going to talk about
one last topic, and then I'll be happy
to respond to questions. So next, another dimension
is whether or not periodontal infection has
an influence on glycemic control in people with diabetes. There are observational
epidemiological studies, and there are also a set of
clinical trials and systematic reviews
and meta-analysis of those clinical trials. Here we can quickly see that by race or ethnicity,
there were greater — there were greater proportions
of people, or the prevalence was higher, in non-Hispanic, Black,
Mexican-American, and also all Hispanics, just as a way of getting
a sense for the burden of
periodontal disease, periodontitis,
in our minority populations. All right, so then, I'm going to
go past this and show you this result here. So the question here
was whether or not the elevated bone loss, a marker of severe periodontitis that you saw early on
in those radiographs, has any influence,
because these people were followed for two years
on whether or not they had — whether or not they had
poorer glycemic control. And in the two younger
age groups — you see the 20-34 and 35-44 — we actually see
significant differences in terms of those in the red box
who had bone loss and compared to those
in the blue bars that did not have bone loss.
And it tells us that for sure, the longitudinal studies,
two-year follow-up, that people who were
at good control at baseline actually developed
poorer control over two-year follow-up than those who had
better controlled diabetes. And with that, I'm going to stop so that we can
entertain some questions. And if we finish early,
I'd like to talk a little bit more about
gestational diabetes, which is another important
part of the story when we think about
relationships between diabetes and oral health. Thank you so much
for your attention so far.
And I'll be asking Gina —
Dr. Thornton-Evans — to tell me about the questions. -Yes, Dr. Taylor.
Thanks so much. One question that came in was with reference to the scale
of the break-off between uncontrolled
and a controlled diabetic. -Yes.
-Actually, in terms of the A1c. -Yes. I showed, you know,
in the Pima Indians study, the cutoff was 9 — 9% for the hemoglobin A1c. That's not the case now.
Typically, the cutoff point
for epidemiologic studies — well, it can vary
in epidemiologic studies. In terms of clinical management, the cutoff now
where we get concerned, or where clinicians
get concerned, is that at anything
7 or greater. -Okay, thank you. Another question, Dr Taylor. Are there any studies
that show impact of treatment of periodontal infection
on glycemic control? -Yes, I can zip to those slides. How many more questions we have? Because I'd love to share that
with you all right now. -Yeah, we have a few more,
but just… -Okay.
-…those slides out. -Let me go to
that set of slides. Okay, so now we're into
another study design. Those are going to be
the randomized controlled style, control studies where we have
a treatment group and we have a control group,
as you can see. We follow them to determine
the differences in outcome. So, yes, there are a series of randomized
controlled trials that I'd like to talk
with you about. So we're talking about non-surgical periodontal
therapy, routine periodontal therapy. So here's an example
of a person. This person could have
had diabetes. These are their lower teeth.
Look how inflamed the tissue is. This is the kind of result
that you can get. See the tissue healing?
Looks so much healthier. And from that, there are a number of systematic
reviews and meta-analyses that have actually
assessed studies. Early on, there were
studies conducted, or systematic reviews, where the number of studies
was 10. These are the number of studies. These are a number of randomized
controlled studies in their review. The type of diabetes,
number of participants, and the change in A1c. This is the most important part. And as long as
the confidence interval doesn't include one, we know that the non-surgical
periodontal therapy, or the routine scaling
and root planing, caused a reduction
in hemoglobin. That would be like going
from 7.4 to 7.0 or going from 8.5
to a reduction to 8.1. So this is what those — these values are,
and the Hemoglobin A1c are, clinically meaningful. They are the equivalent of,
and in some ways, of adding an extra
oral hypoglycemic if a person were taking
oral hypoglycemics. So these are important findings,
and they suggest to us that is a benefit to routine scaling and root
planing in people with diabetes in controlling their glycemia.
Now, with this evidence, in 2013, during that period of time,
NIDCR funded a multicenter
randomized control trial. The population of people with
type 2 diabetes, hemoglobin A1c for inclusion
was 7% to less than 9%. So those — and also, this was the largest clinical trial
that had been conducted. The intervention, of course,
was scaling, root planing, supportive periodontal therapy
at three and six months, and the control group received
no treatment for six months. The outcome was a difference
in hemoglobin A1c change. That was not included
in those systematic reviews that we just looked at
before seeing this. So what were the results? What was the outcome
of that clinical trial? Enrollment stopped early because of futility
of the treatment group. The hemoglobin A1c increased
actually to 0.17% and in the control group,
it increased to 0.11%.
So NIDCR ended the study early because there was no effect. Then there was no significant
difference between the groups. So then, people interested in — colleagues who were interested
in this topic actually noticed some difficulties with —
had difficulties with the way that the study
was designed. First of all, the point for
entry, the inclusion criterion, of 7% to less than 9%,
was reported, and this was published. 7% to 9% — that was too low
to be able to attain an improvement
in glycemic control. Another one is that
the periodontal care failed to reach an accepted
standard of care in terms of resolution.
If you think back
at that picture I showed you of the calculus in the very
inflamed periodontal lower — periodontally diseased
lower anterior teeth, and after treatment,
how the resolution occurred, that was not the kind
of resolution that occurred
in the NIDCR study. So also, there was
no consideration of obesity as another factor in terms of the chronic
inflammatory state. So it was not addressed. So then another set
of systematic reviews came out, and these systematic reviews
included the NIDCR randomized
control trial that I just talked about. The reason that it's significant
that these included the randomized control trial is,
as I've mentioned, that single study actually had
the largest number of — it was a multicenter study,
which is excellent for supporting the results of an RCT randomized control trial. And also, it's important
to include those because the number,
the size of the study, the design of the study,
would have a tremendous weight on the — tremendous impact — and there's ways to calculate
the weight of the impact — on the results of the systematic
reviews and meta-analyses.
As you can see in the change
in hemoglobin A1c, whereas before the NIDCR study
was included in the meta-analysis,
we were seeing 0.4% reduction
in hemoglobin A1c. Now we see it's 0.36,
around 0.3 or so. And yet, this is still a clinically important
reduction. And so it while
you could see the influence of the NIDCR study
that was discontinued and where they found
no significant effect, there still, through
the systematic reviews and meta-analyses, continues
to be strong evidence — that's in my opinion —
strong evidence that routine non-scale
perio-planing — periodontal root scaling
and root planing has a beneficial effect and a clinically meaningful
effect on reduction in hemoglobin A1c.
So why should we be concerned if oral health care
makes a difference? One is that any substantial
lowering of blood glucose in people with diabetes delays the progression
or onset of complications. Another one is every percentage
point reduction in hemoglobin A1c leads to
a reduction in the risk of microvascular
complications of 35% and also a reduction by 0.2%, like we saw in the slides, leads to a 10% reduction
in mortality. And with that, I will stop because I'd like to respond
to a few more questions before we run out of time,
if possible. -Yes, thanks so much,
Dr. Taylor. I just want to just extend
my appreciation for you participating today
in this webinar and for all those
that are actually online and just wanted
to let everyone know that we will have our
next lecture in February 2022.
And Dr. Lois Cohen
will be our speaker, who's an NIH consultant
and also a Paul G. Rogers Ambassador for Global Health
Research at NIDCR. So just check out our website
for more information about that. But one thing I would like
to cover quickly, Dr. Taylor, is you are working
on a policy statement for the American Public
Health Association that focuses on
the PPOD initiative that was initially done by the National Diabetes
Education Program. Could you share very quickly
the work being done with respect to that? -Yeah, and I'm going to
start with the end. The policy statement
was approved. So, and the essence
of the policy statement is to… …is to advocate for
our dental organizations, our pharmacy organizations, our podiatry organizations, as well as
our optemetric organizations to work in collaboration
in a way to enhance care of people
with diabetes.
And so it is to develop the
partnerships and collaborations and to work interprofessionally to be able to upgrade, the level of care,
the level of attention, and also to learn more about
what each person is doing. So it's very closely linked
to the history of PPOD, that's for sure,
very closely linked. So we wrote a policy statement and submitted it to the American
Public Health Association. It was approved last week,
actually, during the APHA meeting.
So that was a work of love
for everyone who was involved. We have —
our oral health component of the American Association
of Dental Public Health had participants, including me. I was one of the participants in
shaping that policy statement. And we are thrilled
to announce that, yes, it was approved. So now the work begins
to build those collaborations and then to actually
show results from those collaborative
interprofessional relationships. -Okay, thank you, Dr. Taylor. We have, I think, a question
from Dr. Timothé… -Okay.
-…from Texas A&M. Dr. Timothé, do you want to
state your question? -I apologize.
I don't have a question.
I'm sorry. I didn't realize
that my computer went on. Hi, Dr. Taylor.
-Hi, Dr. Timothé. -I had no questions.
Thank you. I'm sorry. -Glad to see you.
-No problem. And I think you've pretty much
covered everything just through some of the slides. So that concludes
the webinar for today, and I want to just thank
everyone for participating and for, also,
for people to be aware of our next lecture series that will be in February of 2022 with Dr. Lois Cohen. So thanks to everyone and to our
communication and policy team for all their help with this
particular lecture series today. Thanks..
Leave a Reply