Cover Story

Does periodontal disease cause type 2 diabetes?

Diabetes has long been a risk factor for periodontal disease; new research may point to reverse causation.

Diabetes has long been believed to be a risk factor for periodontal disease. Results of a new study show that the reverse might also be true, according to researchers at Columbia University Mailman School of Public Health. Ryan T. Demmer, PhD, MPH, associate research scientist in the department of epidemiology, said that these findings add a “new twist” to the association, suggesting that periodontal disease may lead to diabetes.

“It has been generally accepted that periodontal disease is a consequence of diabetes despite the fact that this association has not been studied with the same methodological rigor applied to coronary and stroke outcomes,” he told Endocrine Today. “We found that over two decades of follow-up, individuals who had periodontal disease were more likely to develop type 2 diabetes later in life when compared to individuals without periodontal disease.”

The researchers studied over 9,000 participants without diabetes from a nationally representative sample of the U.S. population, with more than 800 eventually developing diabetes. They then compared the risk of developing diabetes over the next 20 years between people with varying degrees of periodontal disease and found that individuals with elevated levels of periodontal disease were nearly twice as likely to become diabetic in that 20-year timeframe. Demmer said to keep an open mind about the results, however.

“They certainly are thought-provoking, biologically plausible and supported by longstanding research regarding periodontal disease and atherosclerotic cardiovascular disease. While there are no immediate clinical implications that stem from these findings, they do suggest a need for additional research,” he said.

Endocrine Today spoke with leading endocrinologists and dentists to discuss how the results of this study may impact the field.

Periodontal disease

Ryan T. Demmer, PhD, MPH
Ryan T. Demmer, PhD, MPH, is an Associate Research Scientist in the Department of Epidemiology at Columbia University Mailman School of Public Health.

Photo by David Wentworth

“Endocrinologists will be a critical component to moving this area of scientific research forward,” Demmer said. He added that the impetus for the current study, which was published in Diabetes Care, was the existing research regarding periodontal infection and CVD.

“The first studies suggesting a link between periodontal disease and both heart attack and stroke were published nearly 20 years ago,” said Moise Desvarieux, MD, PhD, associate professor in the department of epidemiology at Columbia University and senior author of the study. “Since those initial publications, the association has been well established, although it remains unknown whether the association is causal or coincidental.”

Desvarieux and Demmer both noted that the question of whether nondiabetic adults with periodontal disease develop diabetes at a higher rate than those without periodontal disease is one that had not been scientifically tested.

To find answers, they looked at a total of 9,296 men and women without diabetes aged 25 to 74 years who were participating in NHANES I and had completed a baseline dental examination (1971-1976) and at least one follow-up evaluation (1982-1992). They defined six categories of baseline periodontal disease using the periodontal index. Of the 7,168 dentate participants, 47% had periodontal index of zero, which meant they were periodontally healthy. Incident diabetes was defined by death certificate, self-report of diabetes requiring pharmacologic treatment or health care facility stay with diabetes discharge code.

DISCUSS IN OUR FORUM How do you think these findings on diabetes and periodontal disease may impact the field?

Incident diabetes odds were increased by 40% among participants with gingivitis (P0.05) and by 50% among participants with periodontitis (P0.05) compared with periodontally healthy participants. These findings remained after multivariable adjustment for potential confounders including age, smoking, obesity, hypertension and dietary patterns.

Demmer said one hypothesis linking clinical periodontal disease and incident type 2 diabetes involves chronic inflammation resulting from the bacterial infections that often contribute to clinical periodontal diseases.

“It is possible that these bacterial infections could also contribute to chronic elevations in systemic inflammatory mediators (ie, tumor necrosis factor-alpha). Studies have shown that inflammatory mediators such as TNF-alpha can induce insulin resistance, possibly via disruption of signal transduction pathways,” he said. David R. Jacobs, PhD, professor in the division of epidemiology and community health at the School of Public Health, University of Minnesota, and a coauthor of the study, agrees that “inflammation associated with or caused by periodontal disease could lead to diabetes.”

In Desvarieux and Demmer’s study, participants missing 25 to 31 teeth at baseline had an incident diabetes OR of 1.70 relative to participants missing zero to eight teeth (P0.05). Intermediate tooth loss was not associated with incident diabetes.

“This could be suggestive that the people who lost all of their teeth had a history of infection at some point, but subsequently lost their teeth and removed the source of infection,” Desvarieux said.

Chronic inflammation

“These are very interesting data, now adding another risk factor to the long list of risk factors for type 2 diabetes,” said Serge Jabbour, MD, FACP, FACE, associate professor of clinical medicine in the division of endocrinology, diabetes and metabolic diseases at the Jefferson Medical College of Thomas Jefferson University, Philadelphia.

“The results, however, are not completely unexpected, since there was a link between both diseases already discussed in previous studies,” Jabbour said.

Jabbour and Demmer, like most of the experts interviewed by Endocrine Today, said they think the most likely alternative explanation for the observed association is the underlying chronic inflammation behind both diabetes and periodontal disease.

“There is compelling evidence that inflammation and chronic infection play an essential role in the development of type 2 diabetes, and studies in humans suggest that circulating inflammatory marker levels may predict type 2 diabetes years in advance of the onset of these diseases,” said Thomas Beikler, Dr Med, Dr Med Dent Habil, associate professor in the department of periodontics at the University of Washington.

Beikler, who received periodontal specialty training at the University of Muenster, Germany, said the area affected by periodontitis is greater than most professionals would think it is and spans over an area of about 75 cm2, which is equivalent to the palm surface area.

“For that reason it is not astonishing that an increase in local and systemic expression of inflammatory cytokines, such as TNF-alpha and IL-6, has been reported in individuals with periodontitis,” he said. “Both TNF-alpha and IL-6 have been shown to impair intracellular insulin signaling, which may lead to insulin resistance,” said Beilker, who added that these factors may contribute to an increased risk of systemic diseases such as diabetes in patients with periodontal disease.

Karen Earle, MD, medical director of the center for diabetes services at California Pacific Medical Center, said that several possible explanations exist aside from chronic inflammation.

Earle noted the genetic tendency towards both periodontal disease and diabetes. “Another possibility is that patients with poor eating behaviors are more likely to have periodontal disease as well as diabetes. There was some information about diet included in the study criteria — but only through diet recall of the patients.”

Study limitations

Michelle Magee, MD
Michelle Magee

Michelle Magee, MD, director of the MedStar Diabetes Institute, Washington Hospital Center, Washington, DC, and associate professor of medicine, Georgetown University School of Medicine, said that the study is limited by the lack of confirmatory blood glucose measurements for a diagnosis of diabetes at either baseline or at time of follow-up data collection.

“We know that pre-diabetes and indeed type 2 diabetes itself can be present for many years prior to the time of diagnosis,” she said.

“While the investigators have minimized the potential for unrecognized diabetes to have been present by defining a 10-year interval between data samples, it is still possible that fasting glucose or impaired glucose tolerance were present and unrecognized. They may have progressed to type 2 over the course of this time period, particularly during the baseline collection period during the 1970s when there was certainly less awareness of the implications of minimal elevations in blood glucose,” Magee said.

Magee added that if this is the case, the increased incidence of periodontal disease could be attributed to the long-term impact of hyperglycemia on the microvasculature and blood flow to the gums, with resultant impairment of nutrient flow to the gums.

Kashif Munir, MD
Kashif Munir

“This would be compounded by an increased propensity to infection, white cell functional impairment and chronic inflammation in the presence of chronic mild hyperglycemia,” she said.

Kashif Munir, MD, assistant professor of medicine at the University of Maryland School of Medicine, also does not think the study has set anything in stone.

“I don’t think we can say for sure that periodontal disease is causative in this case,” said Munir, “but rather that shared factors such as chronic inflammation or genetics may partially underpin both diabetes and periodontal disease. It would be interesting if these findings could be confirmed, and further if we could test the hypothesis that treating periodontal disease aggressively could decrease the risk of diabetes.”

Reverse causation

Periodontal disease and diabetes are both associated with an increase in pro-inflammatory and a decrease in antiinflammatory cytokines, according to Stuart Weiss, MD, assistant clinical professor of endocrinology at the NYU School of Medicine.

“This is likely how diabetes and periodontal disease are related and how they are seen in association with other conditions, including CVD,” Weiss said.

Endocrinologists need to make patients understand that good dental care is very important and encourage good and aggressive oral hygiene.

“The stress of periodontal disease can lead to a release of chemical mediators that increase inflammation, and the stress of poorly maintained diabetes can make the body more vulnerable to infections and worsening of periodontal disease,” Weiss said. He noted that in his practice he’s seen many cases of diabetes worsened by poor oral health, but with either dental work or more aggressive glucose control — both conditions improve.

Beikler shared a story about interning at a diabetic specialty clinic 16 years ago, before he entered dental school. Beikler said he saw a lot of patients who had difficulty achieving sufficient metabolic control. At the time, he said, no physician would have spent a second thinking about dental problems as a cause for the difficulties in achieving good metabolic control.

Salomon Amar, PhD, DMD
Salomon Amar

“I’m wondering now if some of those patients may have benefitted from routine periodontal examination and treatment,” he said. “I hope that the awareness among all medical and dental professions has increased and that both groups understand that there exists a strong interrelationship between periodontitis and diabetes mellitus that needs a common treatment approach.”

“This reverse causation would imply that endocrinologists, diabetologists and internists would need to refer their patients more often to their periodontist before finalizing their therapeutic approaches, given that treating periodontal disease may favorably affect the control of diabetes,” said

Salomon Amar, PhD, DMD, associate dean for research, Boston University Goldman School of Dental Medicine.

“Diabetic patients are more open to the idea of necessitating a comprehensive periodontal treatment as part of their control of diabetes, but we need more public awareness of the need to consult and refer patients to their dentists to achieve optimal treatment and hopefully early detection,” said Amar, noting that insurance companies have already developed plans for patients with diabetes for coverage of appropriate periodontal therapies.

Next steps

While he is excited about the results, Demmer is the first to admit that researchers should react cautiously to these findings as they require confirmation in other research settings.

“One concern we discussed is the potential for diagnostic bias in which individuals with periodontal disease were possibly more likely to be tested for and diagnosed with diabetes,” Demmer said. “We feel this is unlikely, but it does remain as a possibility.”

Regardless of cause and effect, it is well known that individuals with type 2 diabetes are more likely to have periodontal diseases. Therefore, it seems reasonable for endocrinologists to consider supplying patients with information on the signs/symptoms of periodontal diseases and making referrals to dental colleagues as appropriate.

Demmer and Desvarieux have recently received a grant award from NIH to begin planning a study that will better test this hypothesis. They expect to begin collecting preliminary data soon. They specifically want to test whether colonization with known oral pathogens puts individuals at risk for progression of known diabetes risk factors.

“By collecting information regarding oral bacterial colonization, as opposed to using clinical surrogates of infection, we can focus the hypothesis and get clearer answers,” Demmer said.

Ultimately, if the data continue to demonstrate an association between periodontal infection and diabetes risk, Demmer said, treatment studies will be necessary to demonstrate that periodontal treatment can in fact reduce diabetes risk. – by Angelo Milone

Point/Counter

Do the results of this study have a major impact on the field of endocrinology?

Alan J. Garber, MD, PhD
Alan J. Garber

PERSPECTIVE

Impact of trial design: Perspective from Endocrine Today’s Chief Medical Editor

The periodontal issue is interesting but clearly illustrates the difference from findings resulting from well-designed randomized clinical trials and the findings of epidemiologic, population-based surveys. The former are a higher level of evidence and are accepted proof of causation, whereas the latter show association, not causality. The bases for that association are unclear and may be multiple.

– Alan J. Garber, MD, PhD
Professor in the Departments of Medicine, Biochemistry and Molecular Biology,
and Cellular & Molecular Biology at Baylor College of Medicine, Houston,
and Chief Medical Editor of Endocrine Today

For more information:
  • Demmer RT, Jacobs DR, Desvarieux M. Periodontal disease and incident type 2 diabetes. Diabetes Care. 2008;31:1373-1379.
  • Demmer RT, Desvarieux M. Periodontal infections and cardiovascular disease: The heart of the matter. J Am Dent Assoc. 2006;137:14S-20S.

Diabetes has long been believed to be a risk factor for periodontal disease. Results of a new study show that the reverse might also be true, according to researchers at Columbia University Mailman School of Public Health. Ryan T. Demmer, PhD, MPH, associate research scientist in the department of epidemiology, said that these findings add a “new twist” to the association, suggesting that periodontal disease may lead to diabetes.

“It has been generally accepted that periodontal disease is a consequence of diabetes despite the fact that this association has not been studied with the same methodological rigor applied to coronary and stroke outcomes,” he told Endocrine Today. “We found that over two decades of follow-up, individuals who had periodontal disease were more likely to develop type 2 diabetes later in life when compared to individuals without periodontal disease.”

The researchers studied over 9,000 participants without diabetes from a nationally representative sample of the U.S. population, with more than 800 eventually developing diabetes. They then compared the risk of developing diabetes over the next 20 years between people with varying degrees of periodontal disease and found that individuals with elevated levels of periodontal disease were nearly twice as likely to become diabetic in that 20-year timeframe. Demmer said to keep an open mind about the results, however.

“They certainly are thought-provoking, biologically plausible and supported by longstanding research regarding periodontal disease and atherosclerotic cardiovascular disease. While there are no immediate clinical implications that stem from these findings, they do suggest a need for additional research,” he said.

Endocrine Today spoke with leading endocrinologists and dentists to discuss how the results of this study may impact the field.

Periodontal disease

Ryan T. Demmer, PhD, MPH
Ryan T. Demmer, PhD, MPH, is an Associate Research Scientist in the Department of Epidemiology at Columbia University Mailman School of Public Health.

Photo by David Wentworth

“Endocrinologists will be a critical component to moving this area of scientific research forward,” Demmer said. He added that the impetus for the current study, which was published in Diabetes Care, was the existing research regarding periodontal infection and CVD.

“The first studies suggesting a link between periodontal disease and both heart attack and stroke were published nearly 20 years ago,” said Moise Desvarieux, MD, PhD, associate professor in the department of epidemiology at Columbia University and senior author of the study. “Since those initial publications, the association has been well established, although it remains unknown whether the association is causal or coincidental.”

Desvarieux and Demmer both noted that the question of whether nondiabetic adults with periodontal disease develop diabetes at a higher rate than those without periodontal disease is one that had not been scientifically tested.

To find answers, they looked at a total of 9,296 men and women without diabetes aged 25 to 74 years who were participating in NHANES I and had completed a baseline dental examination (1971-1976) and at least one follow-up evaluation (1982-1992). They defined six categories of baseline periodontal disease using the periodontal index. Of the 7,168 dentate participants, 47% had periodontal index of zero, which meant they were periodontally healthy. Incident diabetes was defined by death certificate, self-report of diabetes requiring pharmacologic treatment or health care facility stay with diabetes discharge code.

DISCUSS IN OUR FORUM How do you think these findings on diabetes and periodontal disease may impact the field?

Incident diabetes odds were increased by 40% among participants with gingivitis (P0.05) and by 50% among participants with periodontitis (P0.05) compared with periodontally healthy participants. These findings remained after multivariable adjustment for potential confounders including age, smoking, obesity, hypertension and dietary patterns.

Demmer said one hypothesis linking clinical periodontal disease and incident type 2 diabetes involves chronic inflammation resulting from the bacterial infections that often contribute to clinical periodontal diseases.

“It is possible that these bacterial infections could also contribute to chronic elevations in systemic inflammatory mediators (ie, tumor necrosis factor-alpha). Studies have shown that inflammatory mediators such as TNF-alpha can induce insulin resistance, possibly via disruption of signal transduction pathways,” he said. David R. Jacobs, PhD, professor in the division of epidemiology and community health at the School of Public Health, University of Minnesota, and a coauthor of the study, agrees that “inflammation associated with or caused by periodontal disease could lead to diabetes.”

In Desvarieux and Demmer’s study, participants missing 25 to 31 teeth at baseline had an incident diabetes OR of 1.70 relative to participants missing zero to eight teeth (P0.05). Intermediate tooth loss was not associated with incident diabetes.

“This could be suggestive that the people who lost all of their teeth had a history of infection at some point, but subsequently lost their teeth and removed the source of infection,” Desvarieux said.

Chronic inflammation

“These are very interesting data, now adding another risk factor to the long list of risk factors for type 2 diabetes,” said Serge Jabbour, MD, FACP, FACE, associate professor of clinical medicine in the division of endocrinology, diabetes and metabolic diseases at the Jefferson Medical College of Thomas Jefferson University, Philadelphia.

“The results, however, are not completely unexpected, since there was a link between both diseases already discussed in previous studies,” Jabbour said.

Jabbour and Demmer, like most of the experts interviewed by Endocrine Today, said they think the most likely alternative explanation for the observed association is the underlying chronic inflammation behind both diabetes and periodontal disease.

“There is compelling evidence that inflammation and chronic infection play an essential role in the development of type 2 diabetes, and studies in humans suggest that circulating inflammatory marker levels may predict type 2 diabetes years in advance of the onset of these diseases,” said Thomas Beikler, Dr Med, Dr Med Dent Habil, associate professor in the department of periodontics at the University of Washington.

Beikler, who received periodontal specialty training at the University of Muenster, Germany, said the area affected by periodontitis is greater than most professionals would think it is and spans over an area of about 75 cm2, which is equivalent to the palm surface area.

“For that reason it is not astonishing that an increase in local and systemic expression of inflammatory cytokines, such as TNF-alpha and IL-6, has been reported in individuals with periodontitis,” he said. “Both TNF-alpha and IL-6 have been shown to impair intracellular insulin signaling, which may lead to insulin resistance,” said Beilker, who added that these factors may contribute to an increased risk of systemic diseases such as diabetes in patients with periodontal disease.

Karen Earle, MD, medical director of the center for diabetes services at California Pacific Medical Center, said that several possible explanations exist aside from chronic inflammation.

Earle noted the genetic tendency towards both periodontal disease and diabetes. “Another possibility is that patients with poor eating behaviors are more likely to have periodontal disease as well as diabetes. There was some information about diet included in the study criteria — but only through diet recall of the patients.”

Study limitations

Michelle Magee, MD
Michelle Magee

Michelle Magee, MD, director of the MedStar Diabetes Institute, Washington Hospital Center, Washington, DC, and associate professor of medicine, Georgetown University School of Medicine, said that the study is limited by the lack of confirmatory blood glucose measurements for a diagnosis of diabetes at either baseline or at time of follow-up data collection.

“We know that pre-diabetes and indeed type 2 diabetes itself can be present for many years prior to the time of diagnosis,” she said.

“While the investigators have minimized the potential for unrecognized diabetes to have been present by defining a 10-year interval between data samples, it is still possible that fasting glucose or impaired glucose tolerance were present and unrecognized. They may have progressed to type 2 over the course of this time period, particularly during the baseline collection period during the 1970s when there was certainly less awareness of the implications of minimal elevations in blood glucose,” Magee said.

Magee added that if this is the case, the increased incidence of periodontal disease could be attributed to the long-term impact of hyperglycemia on the microvasculature and blood flow to the gums, with resultant impairment of nutrient flow to the gums.

Kashif Munir, MD
Kashif Munir

“This would be compounded by an increased propensity to infection, white cell functional impairment and chronic inflammation in the presence of chronic mild hyperglycemia,” she said.

Kashif Munir, MD, assistant professor of medicine at the University of Maryland School of Medicine, also does not think the study has set anything in stone.

“I don’t think we can say for sure that periodontal disease is causative in this case,” said Munir, “but rather that shared factors such as chronic inflammation or genetics may partially underpin both diabetes and periodontal disease. It would be interesting if these findings could be confirmed, and further if we could test the hypothesis that treating periodontal disease aggressively could decrease the risk of diabetes.”

Reverse causation

Periodontal disease and diabetes are both associated with an increase in pro-inflammatory and a decrease in antiinflammatory cytokines, according to Stuart Weiss, MD, assistant clinical professor of endocrinology at the NYU School of Medicine.

“This is likely how diabetes and periodontal disease are related and how they are seen in association with other conditions, including CVD,” Weiss said.

Endocrinologists need to make patients understand that good dental care is very important and encourage good and aggressive oral hygiene.

“The stress of periodontal disease can lead to a release of chemical mediators that increase inflammation, and the stress of poorly maintained diabetes can make the body more vulnerable to infections and worsening of periodontal disease,” Weiss said. He noted that in his practice he’s seen many cases of diabetes worsened by poor oral health, but with either dental work or more aggressive glucose control — both conditions improve.

Beikler shared a story about interning at a diabetic specialty clinic 16 years ago, before he entered dental school. Beikler said he saw a lot of patients who had difficulty achieving sufficient metabolic control. At the time, he said, no physician would have spent a second thinking about dental problems as a cause for the difficulties in achieving good metabolic control.

Salomon Amar, PhD, DMD
Salomon Amar

“I’m wondering now if some of those patients may have benefitted from routine periodontal examination and treatment,” he said. “I hope that the awareness among all medical and dental professions has increased and that both groups understand that there exists a strong interrelationship between periodontitis and diabetes mellitus that needs a common treatment approach.”

“This reverse causation would imply that endocrinologists, diabetologists and internists would need to refer their patients more often to their periodontist before finalizing their therapeutic approaches, given that treating periodontal disease may favorably affect the control of diabetes,” said

Salomon Amar, PhD, DMD, associate dean for research, Boston University Goldman School of Dental Medicine.

“Diabetic patients are more open to the idea of necessitating a comprehensive periodontal treatment as part of their control of diabetes, but we need more public awareness of the need to consult and refer patients to their dentists to achieve optimal treatment and hopefully early detection,” said Amar, noting that insurance companies have already developed plans for patients with diabetes for coverage of appropriate periodontal therapies.

Next steps

While he is excited about the results, Demmer is the first to admit that researchers should react cautiously to these findings as they require confirmation in other research settings.

“One concern we discussed is the potential for diagnostic bias in which individuals with periodontal disease were possibly more likely to be tested for and diagnosed with diabetes,” Demmer said. “We feel this is unlikely, but it does remain as a possibility.”

Regardless of cause and effect, it is well known that individuals with type 2 diabetes are more likely to have periodontal diseases. Therefore, it seems reasonable for endocrinologists to consider supplying patients with information on the signs/symptoms of periodontal diseases and making referrals to dental colleagues as appropriate.

Demmer and Desvarieux have recently received a grant award from NIH to begin planning a study that will better test this hypothesis. They expect to begin collecting preliminary data soon. They specifically want to test whether colonization with known oral pathogens puts individuals at risk for progression of known diabetes risk factors.

“By collecting information regarding oral bacterial colonization, as opposed to using clinical surrogates of infection, we can focus the hypothesis and get clearer answers,” Demmer said.

Ultimately, if the data continue to demonstrate an association between periodontal infection and diabetes risk, Demmer said, treatment studies will be necessary to demonstrate that periodontal treatment can in fact reduce diabetes risk. – by Angelo Milone

Point/Counter

Do the results of this study have a major impact on the field of endocrinology?

Alan J. Garber, MD, PhD
Alan J. Garber

PERSPECTIVE

Impact of trial design: Perspective from Endocrine Today’s Chief Medical Editor

The periodontal issue is interesting but clearly illustrates the difference from findings resulting from well-designed randomized clinical trials and the findings of epidemiologic, population-based surveys. The former are a higher level of evidence and are accepted proof of causation, whereas the latter show association, not causality. The bases for that association are unclear and may be multiple.

– Alan J. Garber, MD, PhD
Professor in the Departments of Medicine, Biochemistry and Molecular Biology,
and Cellular & Molecular Biology at Baylor College of Medicine, Houston,
and Chief Medical Editor of Endocrine Today

For more information:
  • Demmer RT, Jacobs DR, Desvarieux M. Periodontal disease and incident type 2 diabetes. Diabetes Care. 2008;31:1373-1379.
  • Demmer RT, Desvarieux M. Periodontal infections and cardiovascular disease: The heart of the matter. J Am Dent Assoc. 2006;137:14S-20S.