Round 34: Periodontal Disease and Rheumatoid Arthritis

Clifton O. Bingham III, MD
Associate Professor of Medicine
Division of Rheumatology
Department of Medicine
Johns Hopkins University

There are no disclosures relevant to this talk

There will be no discussion of off-label uses of medications, or investigational agents in this presentation

Release Date: July 9, 2010
Expiration Date: January 1, 2011



After completing this activity, you’ll be able to:

  • Discuss Periodontal Disease
  • Explain the Associations of Periodontal Disease and Rheumatoid Arthritis
  • Discuss Citrullination as a shared mechanism of disease


There have been multiple reports that have described an association between periodontal disease and rheumatoid arthritis.  It’s been recognized that periodontal disease is associated with other systemic conditions, including diabetes, atherosclerotic cardiovascular disease, low birth weight infants and preterm labor.  Intervention by treating periodontal disease can actually improve diabetes control, can improve the weight of infants, and can decrease preterm labor.  There is an interventional study that has been ongoing looking at the effect in atherosclerotic cardiovascular disease.  This implies that there may be relationships between oral health and systemic diseases.

Periodontal Disease


  • Inflammatory disease
  • Consequence of an infectious trigger
  • Initially involves gingival soft tissue
  • Characterized by destruction of surrounding connective tissue matrix

Stages of Periodontitis

  • Gingivitis, Chronic inflammation
  • Colonization by periopathogenic “red complex” organisms
  • Loss of connective tissue attachments to teeth
  • Bone resorption
  • Tooth loss

Biofilm Development in PD

Biofilm on the tooth of a patieth with periodontitis.

This is a biofilm on the tooth of a patient with periodontitis – there are multiple bacteria that are piled on top of each other.  Certain bacteria come along and colonize and then there are bacteria that grow on top of them, due to the specific adhesion molecules between the two.  Then additional bacteria that come along with additional properties that build up on top of those.

Colonization in Periodontal Disease

Below is an image of the late colonizers that really describe patients with periodontal disease.

Late Colonizers

Porphyromonas gingivalis

Patients with periodontitis have specific flora in their mouths, of which the most important is Porphyromonas gingivalis.  Others that are also of importance and track along with it, are Treponema denticola, certain species of Eubacterium, Providencia intermedia and actinomycetemcomitans organism here as well.  When these come along the process goes from inflammation to one that has the potential for destruction.

Porphyromonas gingivalis is a gram negative anaerobic bacteria, and it’s one of the terminal “red complex” organisms and a late colonizers.  A property that is unique to P. gingivalis is its expression of LPS.  Most of the other bacteria that are underneath there don’t have LPS—as such, it can activate toll-like receptors.  Also, P. gingivalis has many proteolytic enzymes, many of which are involved in arginine, lysine and cysteine metabolism, including a peptidyl arginine deiminase.  There are also many enzymes that are directly secreted by this bacteria that cause collagen degradation as well as other exotoxins and gingipains which make many proteins susceptible to further cleavage.  P. gingivalis has DNA for enolase, which has overlapping sequence with human alpha enolase, which is susceptible to citrullination.  Antibodies to P. gingivalis are a marker for periodontal disease—one of the ways that people can be identified for having exposure to this organism is by looking at serum antibody responses against it.

Periodontal Disease Epidemiology

  • Leading cause of tooth loss in the US
  • Estimates of Prevalence of Periodontal Disease
    • NHANES III  35% of adults > 35 years
    • 30% of these are moderate to severe (13% total)
    • Substantial proportion of severe PD is progressive
  • Risk factors
    • Cigarette smoking, medications, systemic diseases
  • Twin studies ≈50% heritability
  • Genetic Associations reported
    • DR4, TNF, IL1β, IL6, TLR, CD14 polymorphisms

Healthy/Gingivitis/ Moderate Periodontitis/Severe Periodontitis

Healthy to Severe Periodontitis

The progression of gum disease:

  1. Healthy gums
  2. Gingivitis
    • Some hyperemia of the gingiva
    • A little bit of swelling
    • Some colonization
    • Excessive plaque accumulation—toothbrushing and good oral hygiene will help
  1. Moderate periodontitis
    • Gum inflammation
    • Recession of the gum line—parts of the tooth that weren’t visible now exposed
    • Regression of the gingiva
  1. Severe periodontitis
    • Triangles and black spaces between teeth
    • Significant loss of the gingiva
    • Beginnings of loss of bone
    • Much plaque accumulation

This is a progressive disease and a progressive cycle.

Radiographic Changes in Periodontal Disease

There are many things that are similar between periodontal disease and rheumatoid arthritis, and one of those is the effect on bone.  A normal bitewing x-ray with healthy gums has a normal bone level at the bottom of the tooth.

The bone level in a patient with periodontal disease is regressed significantly and there are spaces underneath the roots of the teeth.

Pathogenesis of Periodontal Disease

  • Bacteria necessary but not sufficient
  • Bacterial eradication does not necessarily lead to resolution
  • Submicrobicidal tetracyclines improve PD (MMP Activity)
  • Antibody response to bacteria predicts progressive disease and bone loss (e.g. IgG Anti- P. gingivalis)
  • T-cell component (HLA, oligoclonal expansions)
  • Cytokine-mediated effector damage (osteoclast, RANKL)
  • Genetic Associations (DR4, IL1 promoter, etc)

Tooth Diagram

Tooth Diagram

This diagram shows what happens in periodontal disease.  The accumulation of bacteria and plaque and biofilm formation that develops in the pocket between the tooth and the gingiva.  These bacteria begin to release substances such as LPS, activating neutrophils, macrophages, and recruiting them into the site.  There are T-cells that proliferate and there are plasma cells in the gingiva as well.  These proliferate and rheumatoid factors have also been described in patients with periodontal disease.  There are also the antibodies that develop again to the bacteria.  The release of multiple cytokines are detectable and are expressed in the gingival tissue as well as released into the gingival crevicular fluid.  There is also activation of fibroblasts that ends up leading to further connective tissue matrix degradation and osteoclast activation that leads to the degradation of bone.  Many of the players that we are seen in rheumatoid arthritis are recapitulated in the periodontal microenvironment.

Previous Studies of RA and PD

NHANES III (dePablo P, et al)

  • Subjects ≥60 years with musculoskeletal and single quadrant dental exams
  • RA more likely edentulous (OR=2.27) and have PD (OR=1.82) compared with non-RA subjects
    • Adjusted for age, sex, race/ethnicity, and smoking

Small case-control studies have reported similar results (Mercado FB, et al; Pischon N, et al)

  • OR 2.3-8.0 for PD in RA patents vs controls

Marotte H, et al

  • High association– periodontal bone loss and wrist X-ray damage (x2=11.82)
  • RA shared epitope associated with both periodontal bone loss and wrist damage (OR=2.2)

Recent RA-PD Case Control Study (Pischon N et al)

  • 57 RA subjects, 52 healthy controls (age and gender-matched)
  • Measurements at 4 sites per tooth, PI, GI, PoD, CAL=CEJ-pocket bottom
  • Periodontally Diseased = mean CAL>4
  • In stepwise logistic regression with RA status, age, gender, education, alcohol and BMI, only RA and age remained predictors of periodontal disease
  • Adjusted OR of 8.05 in RA patients for periodontitis compared with controls (2.93-22.09)
  • When adjusted for PI and GI, OR drops to 6.09 (1.72-21.58)
  • RA patients higher GI, BOP, CAL 4.37 +/- 1.3mm vs 3.4 +/- 0.89 mm , PoD 3.71 +/- 0.73mm vs 3.16 +/- 0.58
  • Limitations: No information of RA disease characteristics, nonstandard definition of disease

RA, PD, and CVD (Abou-Raya S et al)

  • 100 RA patients with CVD, 50 RA patients without CVD, 50 controls without RA
  • 84% of RA patients with CAD, had PD compared to 60% of RA patients without CAD and to 10% of control subjects.
  • BOP, PoD >4mm, AL >4mm, missing teeth, calculus significantly higher in RA patients with CAD as compared to RA patients without CAD and to controls
  • The severity of PD significantly correlated to RA disease duration, DAS28 score, TNF-α level, ESR, hs-CRP, fibrinogen and HDL.

This study gets at this issue of relationships between rheumatoid arthritis, periodontal disease and cardiovascular disease.  Periodontal disease has been described as a risk factor for atherosclerosis and intervention studies have been ongoing trying to look at this relationship better.  There was an additive effect of RA plus cardiovascular disease and the association with periodontal disease, periodontal disease further increasing the cardiovascular risk.

TNF Inhibition Improves Experimental Periodontitis

There have been several studies that have shown that cytokines that are involved in rheumatoid arthritis are also important in periodontal disease.  Several studies have looked at models of experimental periodontitis and shown decreased inflammatory cell infiltration and bone loss with TNF inhibition, including a study that looked at P55 receptor knockout mice that had less bone loss and less severe periodontitis with an induced model of bacteria.

TNF in Patients with Periodontal Disease

TNF has been described as increased in patients with periodontal disease and higher in patients who are smokers than are non-smokers.  A TNF polymorphism has been reported and confirmed recently in Japanese patients with severe periodontal disease.  A very recent study looked at 19 patients with rheumatoid arthritis versus 30 healthy controls and found that the serum TNF level was the only independent variable to predict probing death and attachment loss of bleeding on probing.  It’s a small study with not many controls but it suggests an association with TNF with the extent of periodontal disease.

That was followed, however, with a study in the Journal of Periodontology that showed that infliximab treatment did not improve gingivitis:

  • 40 RA subjects
    • 20 IFX treated (mean 22.2 infusions)
    • 20 with no infliximab
    • No difference in baseline RA disease activity
  • Perio status of IFX patients every 6 weeks
  • 9 subjects from non inflix group with PD (mean probing depth > 4mm) treated with IFX and followed 9 months.
  • Exams on only 6 teeth, max values for each tooth and mean calculated
  • PD and AL not affected by IFX Rx in Group I vs 2
  • Patients with pre/post IFX did have decrease in AL (6.27 +/- 0.97mm vs 5.22 +/- 1.05mm, p<0.05), but PD constant
  • Gingivitis and bleeding index actually increased in 9 patients.
  • Limitations: 6 teeth examined, No control for confounders, Pre-post  only 9 subjects

Shared Mechanisms of Disease in PD and RA

Nitric Oxide
T cells
B cells
Plasma cells

In the general paradigm of RA vs. PD, mediators that are seen in many cell types can be lined up.  Histopathologically, this is a patient with periodontal disease, with an intense neutrophilic, and with mixed cellular infiltrate and even some small lymphoid follicles that appear to be developing in the gingiva, which is very similar to what we might see in a rheumatoid joint.

These relationships that have been described epidemiologically raise the question:  Which comes first, the RA or the periodontal disease?  The traditional paradigm is that RA preceded the development of periodontal disease but that’s far from conclusive—it’s time to begin thinking in the other direction.

Prior Studies of RA and PD

The problems with many of these other studies that have been case-control and otherwise is that they have only looked at patients with established longstanding rheumatoid arthritis.  There has been very poor characterization of RA disease activity.  The influence of medications and other comorbid medical conditions has been unclear and uncontrolled.  There has been poor characterization of any type of functional ability for the RA patients, especially hand function.  Also, there has been no control for Sjögren’s, and they have been poorly controlled for smoking.

Old vs. New Hypotheses

The old hypothesis:  RA is Risk Factor for Periodontal Disease.  Patients with RA had poor oral hygiene because they couldn’t grip their toothbrush and brush well, or they had RA and had TMJ joints that were affected and couldn’t open their mouth enough to be able to brush their teeth.  Or they were immunosuppressed or received NSAIDs that could be risk factors for the development of periodontal disease, or they had secondary Sjögren’s.  None of these were proven in any way, however.

The new hypothesis:  Periodontal Disease is a Risk Factor for RA.  There are similar genetic susceptibilities, in terms of HLA-DR4 by two studies and some of these other polymorphisms that have been described.  There are similar lesions in the inflammatory focus and we there are other bacterial infections that can directly cause inflammatory arthritis, either by molecular mimicry or by other mechanisms, including rheumatic fever, Lyme disease and reactive arthritis.

Question #1 – Do patients with Early RA have increased periodontal disease?

Periodontal Assessment in RA Study (PARAS, R03 DE018094)

  • Inclusion
    • 18-75 yrs
    • Definite or probable RA
    • RA diagnosis within 12 months
    • At least 20 teeth
  • Exclusion
    • Other inflammatory arthritis
    • Systemic or oral topical antibiotics within 30 days
    • Phenytoin, cyclosporine, coumadin, or other medications associated with gingival hyperplasia
    • Contraindications to probing (e.g. coagulopathy)
    • Valvular heart disease, prosthesis, or other conditions requiring ABX prophylaxis for dental procedures (unless willing to receive standard prophylaxis)

RA Characteristics Explored

RA Severity and Extra-articular Factors

RA Disease Activity

RA Treatment


  • RA duration
  • RF
  • Anti-CCP
  • Nodules
  • Damage
  • Ro/La
  • Schirmer test
  • DAS28
  • Swollen joints
  • Tender joints
  • MD global
  • Pt Global
  • HAQ
  • CRP
  • Functional Class
  • DMARDs
  • Prednisone
  • NSAIDs
  • Demographic
  • Smoking
  • Alcohol
  • Aspirin use
  • Brushing
  • Flossing
  • Diabetes
  • Concom Meds
  • Medical Hx

Dental Characteristics Explored

Dental History

Periodontal Questions

Sjogren’s Questions

Dental Exam

  • Dental History
  • Tooth loss
  • Hygiene
  • Brushing
  • Flossing
  • Smoking
  • Alcohol
  • Bleeding Gums
  • Swollen Gums
  • Halitosis
  • Pockets
  • Gaps between teeth
  • Oral Dryness
  • Ocular Dryness
  • Overall state of dentition
  • Examine all teeth
  • 6 sites per tooth
  • Multiple measures
  • Unstimulated salivary flow and saliva collection

Periodontal Examination

The periodontal examination being performed is an overall oral assessment.  All teeth except the wisdom teeth are examined at six sites on every tooth for the following indices:  plaque, gingivitis, pocket depth, distance from the edge of the gum down to where the bottom of the pocket is, the amount of recession or attachment loss, bleeding on probing, tooth mobility and furcation (sticking the probe in and seeing if it can get underneath a root).

Examples of Healthy/Gingivitis/Moderate Periodontitis/Severe Periodontitis

Examples of Periodontal Progression


  • Periodontally Healthy
    • At least 24 teeth
    • No attachment loss > 3mm
  • Periodontally Diseased
    • At least 20 teeth
    • At least 4 sites with pocket depths greater than 3mmand
    • At least 4 sites with attachment loss greater than 3mm
  • Periodontal Progression
    • More than 2mm loss of attachment from baseline to 6- month measurement at more than 4 sites

Oral Health Assessment in RA Patients

Assessment of the first 26 patients that were evaluated, breaking down patients between those with early rheumatoid arthritis and those with established rheumatoid arthritis showed that the difference in disease duration is shorter disease in the patients with early RA.  The CRPs were a bit lower in the group of patients with early RA probably because there is a little bit more steroid use in these patients.  The tender joint counts were higher in the early rheumatoid arthritis patients.  The DAS values were not statistically significantly different, but the numbers of patients remained somewhat low with this preliminary data.
PD is Equally Present And Severe Early And Late In The Course of RA
For the periodontal parameters, the numbers of teeth were similar.  With regards to the number of patients who met our definition for periodontally diseased:  65.4 percent overall, 54.5 percent in the early and 73.3 percent in the established.  Based on the NHANES III definition, the expected value in the population would be about 35 percent.


  • Periodontal disease is both prevalent and often severe in patients with RA
  • Characteristics of periodontal disease are similar in patients with early and established disease

Question #2:  Do patients with RA and periodontal disease have more severe RA?


  • Periodontal disease and oral symptoms will be common in RA patients
  • RA disease activity and severity will be associated with periodontal disease


  • To estimate the prevalence of periodontal disease in patients with RA using self-report oral health questionnaires
  • To explore the associations of RA disease characteristics with periodontal and oral health measures in RA patients


The method used was to administer questionnaires asking about periodontal disease to patients to get data concerning the relationships with rheumatoid arthritis.  The available population of patients with the ESCAPE cohort were administered questionnaires regarding periodontal disease and oral health.  (ESCAPE RA is the Evaluation of Subclinical Atherosclerosis and Predictors of Events in RA.)  183 patients from the ESCAPE cohort took the questionnaire.  RA disease characteristics and their oral health parameters were analyzed, and a validation study in the 33 patients that evaluated with oral examinations and questionnaires was also performed.  ESCAPE patients are aged 45 to 84 and are recruited from the Johns Hopkins Arthritis Center and community rheumatologists.  Patients with prior physician-diagnosed cardiovascular events or procedures are excluded.

Methods: Analysis

  • Univariate associations for RA disease characteristics with periodontal disease characteristics
  • Poisson regression of periodontal symptoms with RA characteristics with robust variance estimation
  • Prevalence ratios (PR) calculated in simple and multivariable models
    • adjustments for socio-demographic, lifestyle, oral hygiene, and co-morbid disease variables
  • Association of periodontal and oral/ocular dryness symptoms with log transformed DAS28 score modeled using multivariable linear regression in zero order and complex models
  • Simpler multivariable models constructed by comparing Akaike’s Information Criteria (AIC) values for complex vs. simpler nested models
  • Shapiro-Wilk test to examine normality across the extent of independent variables in multivariable modeling
  • Adjusted coefficient of determination (R2) calculated for each model to represent the proportion of the variability in outcome explained by aggregate of modeled covariates

Characteristics Explored

RA Severity
and Extra-articular Factors

RA Disease Activity

RA Treatment


  • RA duration
  • RF
  • Anti-CCP
  • Shared epitope
  • Sharp score
  • Nodules
  • Ro/La
  • Schirmer test
  • DAS28
  • Swollen joints
  • Tender joints
  • MD global
  • HAQ
  • CRP
  • DMARDs
  • Prednisone
  • NSAIDs
  • Smoking
  • Alcohol
  • Aspirin use
  • Brushing
  • Flossing
  • Diabetes

RA Disease Characteristics (n=183)

Age, years

61.5 +/- 8.3

Female (percentage)


RA Duration yrs, median (IQR)

10.8 (5.5-18.8)

RF-positive (percentage)


CCP-positive (percentage)


HLA-DRB1 Shared epitope (percentage)


Nodules (percentage)


Swollen 28 Joint Count, median (IQR)

3 (1-5)

Tender 28 Joint Count, median (IQR)

3 (1-6)

CRP, mg/L, median (IQR)

2.75 (0.95-7.24)

DAS28 CRP, median (IQR)

3.12 (2.43-3.79)

HAQ, units, median (IQR)

0.75 (0.13-0.15)

RA patients have longstanding disease, which is relatively well-controlled.

Oral Health Characteristics

Current smoking (percentage)


Any periodontal symptoms (percentage)


Receding gums (percentage)


Bleeding gums with brushing (percentage)


Bleeding gums without brushing (percentage)


Swollen gums (percentage)


Bleeding OR swollen gums (percentage)


Bleeding AND swollen gums (percentage)


Oral dryness (percentage)


Ocular dryness (percentage)


Schirmer positive (percentage)


Ro or La positive (percentage)


Self-reported oral symptoms are common in RA patients.

Univariate Associations Between RA Characteristics and Gum Bleeding/Swelling
Univariate associations between the RA characteristics and whether or not patients had gum bleeding or gum swelling were examined..  Nodules, disease activity, swollen joint count, tender joint count, HAQ and physician global were all statistically significantly associated with patients’ self-report of periodontal symptoms.  In bivariate analysis the only RA characteristics that remained significantly associated with self-reported periodontal symptoms, swollen gums, or bleeding gums were DAS28 and rheumatoid nodules.

Associations of RA Disease Activity (DAS28) with Periodontal Symptoms
The only characteristics that remained associated after controls, however, were rheumatoid nodules in DAS28.  Looking further into the relationship between DAS28, with any periodontal symptoms, with swollen gums and with bleeding gums and breaking out DAS28 levels by patients with high disease activity,  moderate disease activity and low disease activity, there is  what looks like a dose-response relationship between the presence of periodontal symptoms and disease activity with RA.  This data was adjusted for gender, race, age, education, smoking, alcohol, brushing, flossing and aspirin use.  All of the other variables (diabetes, family history of bleeding disorder, RA duration, HAQ, RF, anti-CCP, Ro, La, nodules, Sharp score, cumulative prednisone dose, and HLA-DRB1 “Shared Epitope”) were also examined and none of them were significant.

DAS28 Scores According to Combined Gum Bleeding, Swelling, and Oral Dryness

DAS28 Scores

It appears that patient report of oral dryness further increases the disease activity for rheumatoid arthritis.  These are patients reporting whether or not they had these periodontal symptoms of bleeding, swelling of their gums and now reporting whether or not they had dryness in their mouth. There appeared to be an additive effect on top of the effect from the bleeding and swelling in terms of relationship with RA disease activity.
Periodontal Symptoms Account for 22% of variability in DAS28

  • Despite adjusting for confounders, gum bleeding and swelling remained significantly associated with higher log DAS28 scores.
  • Proportion of the variability in log DAS28 score explained by the full model was 33% (i.e. adjusted R2=0.326), reduced to 25% when periodontal symptoms were excluded from the model.
  • Difference in explainable variability in log DAS28 attributed to the independent association of gum bleeding and swelling (22% of the total explainable variability) was highly significant (p<0.001).

Validation of Periodontal Self-Report

  • 26 RA subjects had comprehensive full mouth oral examinations with questionnaires administered
  • The odds of having PD* in those reporting periodontal symptoms was 2.3 times higher than those without symptoms
  • Performance of questionnaires
    • Specificity: 0.75
    • Sensitivity: 0.40
    • Positive predictive value: 0.595
  • High specificity of individual questions for bleeding/swelling with bleeding on probing and gingivitis
  • Questionnaires, while not detecting all cases, provide some indication of a true association

* – PD defined as Pocket depth ≥3mm at ≥4 sites, and Attachment loss ≥3mm at ≥4 sites

Self Reported Gum Bleeding with Examination
There is a statistically significant relationship between these.

Correlation of Gingival Bleeding with Periodontitis
Looking at the correlation between gingival bleeding and periodontitis there relationship that is highly statistically significantly different.  Even though periodontal examinations were not performed there is some evidence that the questionnaires that were administered do show a relationship with the ultimate question of periodontal disease being there or not.
Limitations of Study

  • Questionnaires and self-report
  • Patients with longstanding RA
  • Relatively well-controlled RA disease

Conclusions of this Study

  • In this large cohort of established RA patients, a high prevalence of self-reported periodontal symptoms was observed, consistent with our earlier report based on oral examination
  • Questionnaires had a moderately high positive predictive value to detect periodontal disease
  • Periodontal symptoms were associated with increasing RA disease activity
    • Even after adjusting for multiple pertinent confounders
  • Oral dryness was an additive factor associated with RA disease activity

Additional Questions, and Citrullination

  • Do patients with RA have laboratory evidence of exposure to oral pathogens?
  • Do patients with periodontal disease have antibodies associated with RA?
  • Can careful attention to oral health improve outcomes for patients with RA?
  • Does periodontal disease contribute to increased cardiac risk in RA patients?

P. gingivalis is a unique organism in the mouth in that it has an endogenous peptidylarginine deiminase (PAD) enzyme.  Citrullination is one of the critical first steps in the development of rheumatoid arthritis where markers are detected as antibodies against citrullinated peptides, a disease specific marker that also correlates with disease and with damage.

Citrullination is the conversion of a charged arginine residue to an uncharged citrulline residue, which leads to secondary and tertiary conformational changes in proteins, potentially forming neoepitopes.  In rheumatoid arthritis numerous putative targets of citrullination have been identified, including fibrinogen, vimentin, enolase and even PAD4.  (There are five PAD4s in humans.)  PAD2, PAD4 and PAD6 are expressed in RA synovium and PAD4 polymorphisms have been described in some RA patients.  Recently it has been learned that PAD2 is upregulated with cigarette smoking and also that P. gingivalis has an endogenous PAD.

P. gingivalis has an endogenous PAD because one of the things that happens through this conversion of arginine to citrulline is the release of ammonia.  In the mouth, one of the NA immune defenses is acidification of the environment of the mouth, which is typically inhospitable to bacteria.  In the microenvironment at the base of the tooth it would be very advantageous for a bacteria to be able to neutralize acid.  A bacteria that is able to citrullinate proteins that are readily available, therefore generating ammonia, can provide for itself a more hospitable environment for it to evade host defense.  The process of citrullination, at least through human PADs, can also inactivate certain chemokines that could lead to neutrophil recruitment into the area.  Thus, two potential mechanisms for host defense evasion from the bacteria through a PAD.

Roles of Periodontitis and PADs in Citrullination

Roles of Periodontitis and PADs in Citrullination

Now, throw periodontitis into the mix.  For example, smoking is potentially a trigger for turning on human PAD, such as PAD2, activating and releasing IL1 and TNF, which may also act to stimulate human PADs and periodontitis acting through LPS and toll-like receptors, which could potentially have an additional impact on human PAD regulation.  P. gingivalis specifically has its own peptidylarginine deiminase and also has gingipains, which make many of these proteins potentially more susceptible to cleavage by other proteases.  There is a whole panel of different putative citrullinated substrates that have been implicated for rheumatoid arthritis as well.

It has also been shown that human enolase, alpha enolase, is a putative autoantigen for patients with rheumatoid arthritis.  The site that becomes citrullinated in this human enolase has shared sequence hemology with bacterial enolase that is made by P. gingivalis.  The response to and generation of autoantigens against these enolase products are very highly MHC restricted, not just within the shared epitope but with specific ileals.  This indicates that there may be a very fine specificity between what becomes citrullinated and what becomes neoantigenic in specific individuals.  This is possibly where, based on this hypothesis, that periodontitis could come into play.

Citrullinated Proteins are Detected in Saliva from Patients with Systemic Autoimmune Diseases

The question then becomes do we see any evidence for citrullination in patients with periodontal disease and in patients with rheumatoid arthritis in the mouth?

Logistic Regression for Periodontal Disease with RA Parameters (n=26)


Female Gender

Current Smoking



Perio R03 Definition





CDC 2007 Moderate





CDC 2007 Severe





  • In logistic regression, cigarette smoking, anti-CCP status, RA disease activity (DAS28), and female gender were associated with increased odds for PD using various definitions
  • Anti-CCP with PD and smoking with severe and moderate PD reached or approached significance

This is saliva just collected from patients with RA, psoriatic arthritis and Sjögren’s syndrome run out on a blot and then probed with the AMC citrullination detection kit.  It is a positive control of neutrophils stimulated with calcium and ionomycin.  Saliva is from patients with various autoimmune conditions showing citrullinated substrates across the board.  Citrullinated proteins are detected in the saliva, in some more than others, the identity of which are not clear.  If protease inhibitors aren’t put in immediately after the saliva is obtained, much of it will be lost.  The procedure for saliva collection has been modified to include a cocktail of protease inhibitors to try to stabilize what being studied so it can be evaluated further in the future.

IgG Anti-CCP Antibodies are Associated with Antibody responses to P. gingivalis

  • Random sample of RA database patient serum
  • 76 patients equally distributed between CCP-positive and CCP-negative
  • Serum batch shipped, without reference to CCP state, for IgG antibody against cell surface antigens for P gingivalis using a previously established ELISA.
  • Results: Odds Ratio of 2.67 (CI 1.04-6.8) for anti-CCP with anti-P gingivalis antibodies.
  • These results are consistent with a recent report by Mikuls who showed arelationship with IgM anti-CCP and IgG2 anti-CCP (but not IgA or total IgG) with anti-P gingivalis.
  • Taken together these studies show consistent associations between periodontal bacteria and antibody responses against citrullinated peptides.

PAD-4 is Detected in Periodontal Tissue

  • PAD4 expression was examined using immunohistochemistry of gingival tissue from patients with periodontal disease undergoing surgery.
  • Preliminary Results: Nuclear and cytoplasmic expression of PAD4 was seen in periodontitis predominantly in nonkeratinized junctional and pocket epithelium adjacent to the gingival sulcus; the level of expression was somewhat less than seen in other inflammatory tissues such as RA synovium.


  • Moderate to severe periodontal disease is common in RA patients at disease initiation and with established disease
  • RA patients with more periodontal symptoms have higher levels of RA disease activity
  • In RA patients, PD is a strong predictor of anti-CCP antibodies
  • Antibodies indicating exposure to an oral bacteria with the ability to citrullinate, P. gingivalis, are associated with anti-CCP antibodies.

Further Studies

  • Comprehensive oral assessments of RA patients and matched controls are underway to further confirm these self-report findings
  • Evaluations of the mechanistic underpinnings of disease interrelationships are also in progress
  • Whether aggressive management of periodontal disease in RA patients will result in improved RA clinical outcomes remains to be determined

Clinical Studies

  • Continued longitudinal evaluation of early RA patients with sequential periodontal exams
  • Oral examinations of ESCAPE patients for validation of questionnaire data and additional associations
  • Effects of RA medications on PD parameters
  • Further evaluation of Sjogren’s and relationships with PD and RA
  • Analysis of additional questionnaire data from BRASS cohort
  • Gingival tissue collection of patients undergoing periodontal surgery
  • Interventional studies to determine the effect of aggressive treatment of periodontal disease on RA outcomes

Current RA Patient Recruitment for Oral Health Assessment

  • Screened: 97
  • Enrolled: 56
  • First Exam completed: 38
  • Second Exam completed: 10

Treatment of PD Improves RA Disease Activity (Ortiz P, et al)

A very recent treatment study of PD in RA patients just came out which took 40 patients with RA, all of whom had generalized severe periodontitis.  (It does speak again to the prevalence of severe periodontal disease in RA patients if they could identify 40 patients over a year and a half to do the study.)  Twenty of the patients were on DMARDS and twenty were on DMARDS plus TNF.  In each group the patients were randomized either to get treatment for their periodontal disease or to wait six weeks, and then the periodontal exam was performed again.  There was a reduction in SED rate in DAS28, in periodontal parameters with the periodontal treatment over six weeks.  Interestingly, there was a reduction from baseline TNF levels in 40% of the patients who received the periodontal treatment versus 20% of those who had not been treated, even in the patients receiving TNF antagonists.  This is very intriguing data—it’s a small study, but it does hint at the possibility of doing an interventional study looking at RA with periodontal disease.

Triangulation of Inflammation

Triangulation of Inflammation

Mechanistic Studies

  • Bacterial PAD
    • Cloning, sequencing, expression
    • Functional analysis
    • Calcium and pH dependence
    • Substrate specificity
    • Comparison with human PADs
  • Gingival tissue, gingival crevicular fluid, and salivary expression
    • Citrullinated proteins, PADs, relation to inflammation
    • Role of citrullination in periodontal disease and Sjogren’s
    • PD +/- RA, Healthy (no-PD, non-RA)‏
  • Antibodies to bacterial flora vs colonization patterns
    • Examination of RA, PsA, non-arthritis controls
    • Relation to anti-CCP and RA parameters
  • Relationship of bacterial colonization, PD, and SE status


  • Periodontal disease is prevalent and often severe in patients with RA
  • Significant evidence suggests that citrullination may link periodontal disease with RA
  • Additional studies are needed to better understand these mechanisms

Clinical Implications

  • Oral health parameters should be more closely monitored in patients with RA and autoimmune diseases
  • Interventions to improve oral pathology may have direct and indirect systemic benefits


All information contained within the Johns Hopkins Arthritis Center website is intended for educational purposes only. Physicians and other health care professionals are encouraged to consult other sources and confirm the information contained within this site. Consumers should never disregard medical advice or delay in seeking it because of something they may have read on this website.

Updated: July 31, 2012

Arthritis Center

About Arthritis Center

Founded in 1998, the Arthritis Center at Johns Hopkins is dedicated to providing quality education to patients and healthcare providers alike.