Periodontitis and Rheumatoid Arthritis Review

Document Sample
Periodontitis and Rheumatoid Arthritis Review Powered By Docstoc
					                                                                                                                Volume 76 • Number 11 (Suppl.)

  Periodontitis and Rheumatoid Arthritis:
  A Review
  P.M. Bartold,* R.I. Marshall,† and D.R. Haynes‡

       Periodontitis and rheumatoid arthritis (RA) appear to share
       many pathologic features. In this review, the common patho-
       logic mechanisms of these two common chronic conditions
       are explored. Emerging evidence now suggests a strong rela-
       tionship between the extent and severity of periodontal disease
       and RA. While this relationship is unlikely to be causal, it is clear
       that individuals with advanced RA are more likely to experience
                                                                                                ‘‘In fact, adult periodontitis and
       more significant periodontal problems compared to their non-
                                                                                                rheumatoid arthritis have much in
       RA counterparts, and vice versa. A case is made that these
                                                                                                common, so much so that I have
       two diseases could be very closely related through common un-
                                                                                                argued that they are really the same
       derlying dysfunction of fundamental inflammatory mecha-
       nisms. The nature of such dysfunction is still unknown.
       Nonetheless, there is accruing evidence to support the notion

                                                                                                     he above bold and challenging
       that both conditions manifest as a result of an imbalance be-                                 statement may seem to be
       tween proinflammatory and anti-inflammatory cytokines. As                                       stretching the boundaries of con-
       a result, new treatment strategies are expected to emerge for                          ventional thought too far. However,
       both diseases that may target the inhibition of proinflammatory                         close inspection of two of the most com-
       cytokines and destructive proteases. The clinical implications                         mon chronic diseases afflicting humans
       of the current data dictate that patients with RA should be care-                      reveals remarkable similarities that war-
       fully screened for their periodontal status. J Periodontol                             rant further investigation.
       2005;76:2066-2074.                                                                        The relationship between rheumatoid
       KEY WORDS                                                                              arthritis (RA) and the progression of in-
                                                                                              flammatory conditions elsewhere in the
       Arthritis, rheumatoid; periodontitis.
                                                                                              body, such as periodontitis, is controver-
                                                                                              sial. While a number of studies have
   * Department of Dentistry, University of Adelaide, Adelaide, South Australia, Australia.
   † Department of Dentistry, University of Queensland, Brisbane, Queensland, Australia.
                                                                                              presented conflicting results regarding
   ‡ Department of Pathology, University of Adelaide.                                         a relationship between periodontitis and
                                                                                              RA, there have been recent reports sug-
                                                                                              gesting a significant association between
                                                                                              these two common chronic inflammatory
                                                                                              conditions.2-6 In light of these reports,
                                                                                              there is a need for further investigations
                                                                                              to determine whether the severity of RA
                                                                                              and the severity of periodontitis are inter-
                                                                                              related. To do this, controlled, population-
                                                                                              based, laboratory, and clinical (molecular
                                                                                              epidemiology) studies are needed to ver-
                                                                                              ify the immunological and biological as-
                                                                                              sociations between RA and periodontal
                                                                                              PERIODONTAL DISEASES
                                                                                              The periodontal diseases range from the
                                                                                              relatively benign form of gingivitis to ag-
                                                                                              gressive periodontitis. Many of these
                                                                                              conditions are not only a threat to the
                                                                                              dentition, but may also be a threat to

J Periodontol • November 2005 (Suppl.)                                                         Bartold, Marshall, Haynes

general health. There are reports suggesting in-              From the natural history studies of RA and peri-
creased prevalence of diabetes, atherosclerosis,           odontitis, it has been observed that certain RA and
myocardial infarction, and stroke in patients with peri-   periodontitis populations are characterized by a
odontal disease.7-9 Thus, the likelihood of periodontal    particular type of patient who will experience dis-
disease being associated with systemic diseases is be-     ease progression irrespective of any treatment pro-
coming established fact. All forms of inflammatory          vided. Whether the RA group, in which disease
periodontal disease are associated with chronic in-        progression seems uncontrolled even after compre-
flammation (accumulation of B and T lymphocytes             hensive treatment, is the same group that is suscepti-
as well as monocytes and neutrophils), resulting in de-    ble to develop severe forms of periodontal disease
struction of the periodontal ligament and bone. If left    remains to be established and is, indeed, a major
untreated, significant tissue damage occurs, and the        thesis of this review.
affected teeth can become loose and may be lost if            Rheumatoid arthritis. At least three types of dis-
the disease continues to be active. What is particularly   ease manifestation can also be observed in RA pop-
curious about this disease is the great variability in     ulations: 1) self-limited: in these cases individuals
presentation. Because of its multifactorial nature,        originally presenting for RA have no evidence of
which is modified by systemic, environmental, and           disease 3 to 5 years later;12 2) easily controlled: the
microbiological factors, not all individuals are af-       disease is relatively easily controlled with only non-
fected to the same degree despite the ubiquitous pres-     steroidal anti-inflammatory drugs (NSAIDs);13 3) pro-
ence of dental plaque.                                     gressive: these patients generally require second-line
                                                           drugs, which often still do not fully control the dis-
RHEUMATOID ARTHRITIS                                       ease.14,15
Rheumatoid arthritis is also a chronic destructive in-
flammatory disease10 characterized by the accumula-         Etiologic Factors
tion and persistence of an inflammatory infiltrate in           Periodontitis. The periodontal diseases are well
the synovial membrane that leads to synovitis and          recognized as classic examples of chronic inflamma-
the destruction of the joint architecture resulting in     tory diseases resulting from the induction of host in-
impaired function. As a systemic disease, RA has           flammatory responses to the subgingival biofilm.
extra-articular manifestations in systems such as the         Gingivitis is typically characterized as a robust in-
pulmonary, ocular, vascular, and other organs or           flammatory response confined mainly to the superfi-
structures that may be affected by the inflammatory         cial gingival connective tissues and is a relatively
process. The current paradigm for RA includes an ini-      nonspecific response to a nonspecific accumulation
tiating event (possibly a microbial exposure or a puta-    of dental plaque. How gingivitis progresses to peri-
tive autoantigen) leading to significant synovial           odontitis is still unclear.16
inflammation and tissue destruction. As for periodon-          Periodontitis, on the other hand, appears to be
titis, there is an accumulation of inflammatory cells (T    a more specific inflammatory response to specific
and B lymphocytes, neutrophils, and monocytes), tis-       periodontal pathogens residing in the subgingival bio-
sue edema, endothelial cell proliferation, and matrix      film. Within the conditions known as periodontitis,
degradation. RA is also modified by systemic, genetic,      there is considerable variability in terms of clinical
and environmental variables.                               manifestation and disease progression rates. This var-
                                                           iability can be attributed to differences in composition
SIMILARITIES BETWEEN RHEUMATOID                            of the subgingival microbial flora, as well as factors
ARTHRITIS AND PERIODONTITIS                                that modify the host response to the microbial chal-
Natural History                                            lenge. Nonetheless, it must be noted that, although
   Periodontal disease. Natural history studies of         bacteria are necessary for disease initiation, they
periodontal disease in humans indicate the presence        are not sufficient to cause disease progression unless
of three distinct subpopulations: 11 1) no progression     there is an associated inflammatory response within
of periodontal disease, in which around 10% of the         a susceptible host.16
population manifest very little or no disease which is        Rheumatoid arthritis. Although the cause of RA is
of no particular consequence to the dentition; 2) mod-     unknown, it has been recognized that many different
erate progression, affecting around 80% of the popu-       arthritogenic stimuli activate inflammatory responses
lation and representing a very slowly progressing form     in immunogenetically susceptible hosts. Thus, stud-
of disease that generally can be easily managed via        ies have focused on exogenous infectious agents, en-
routine therapies; and 3) rapid progression, affecting     dogenous substances, such as connective tissue
approximately 8% of individuals whereby extensive          proteins (e.g., collagens and proteoglycans), and al-
periodontal destruction occurs which can be very dif-      tered immunoglobulins as the causative candidates.
ficult to control.                                          The concept that RA is an infectious disease has been

Periodontitis and Rheumatoid Arthritis                                                         Volume 76 • Number 11 (Suppl.)

   considered for over 70 years.17 The idea that RA pa-         odontitis, there is a report indicating that it is an
   tients acquire an infection that elicits an immune re-       important component of the genetic susceptibility to
   sponse in the synovial membrane would account for            some forms of this disease.22 In addition, polymor-
   some of the clinical features and would also explain         phisms in the interleukin-1b (IL-1b) gene cluster have
   the accumulation of immunocompetent T and B cells            been shown to have a significant correlation with
   in the lesions. Data from different animal models dem-       some forms of periodontitis in certain populations.23
   onstrate that arthritis can develop secondarily to sev-         Rheumatoid Arthritis. Family studies and studies
   eral different stimuli and through several different         on monozygotic and dizygotic twins have shown
   effector pathways. If such observations are also appli-      that RA has several features indicative of a complex
   cable to human RA, we might anticipate that different        genetic disease including genetic variance, incom-
   types of infections as well as other environmental           plete penetrance, and multiple gene involvement.24
   exposures with capacity to induce excessive proin-           For RA, the strongest genetic associations are found
   flammatory cytokines in genetically susceptible indi-         within the HLA genes.23 Using DNA sequencing and
   viduals may all potentially contribute to disease either     molecular-based typing, it has been demonstrated
   in unison or isolation.                                      that the disease-conferring portion of the D region is
                                                                confined to a short sequence within the third hyper-
   Are Bacteria a Common Etiologic Link Between
                                                                variable region of HLA-DRB1 gene which includes
   Periodontitis and Rheumatoid Arthritis?
                                                                the amino acid positions 67 through 74.25,26 The
   There are a number of shared features between
                                                                HLA genes and gender constitute about 30% of the ge-
   microorganisms that can induce RA in a genetically
                                                                netic risk in RA, while other genetic factors such as
   susceptible host and the recognized periodontal
                                                                cytokine genes, germline genes, and T-cell receptors
   pathogens. Nonetheless, RA is still not largely recog-
                                                                also account for some of the genetic predisposition
   nized as a disease resulting solely from bacterial
                                                                to RA.10
   challenge. On the other hand, technological and con-
   ceptual advances have permitted the identification of
                                                                Effector Mechanisms of Tissue Destruction in
   bacteria or groups of bacteria associated with specific
                                                                Rheumatoid Arthritis and Periodontitis
   periodontal diseases.18 Close inspection of the viru-
                                                                There is almost universal acceptance that a variety of
   lence factors of periodontal pathogens would suggest
                                                                cytokines and matrix metalloproteinases (MMPs) are
   that such a response could be feasible. Thus, the pos-
                                                                upregulated and intimately involved in the pathogen-
   sibility that ongoing periodontitis could trigger RA in
                                                                esis of both periodontitis and RA; many of these effec-
   genetically susceptible individuals is plausible.
                                                                tor molecules appear to be common to both diseases.
      Notwithstanding the above, these concepts remain
                                                                The task now is to identify the specific cytokines, their
   speculative until the causative agent for RA can be de-
                                                                concentrations, the cells they affect in vivo, the stages
   finitively identified. To date, no infectious agents have
                                                                in which they are active, and the role and concentra-
   been identified as the cause of RA in humans. Indeed,
                                                                tions of their inhibitors. While the effects of cytokines
   current information does not support the concept that
                                                                on normal cellular process are important, it is their
   a single antigen is responsible for synovial inflamma-
                                                                purported roles in pathophysiology that may result
   tion. It is possible that there is no single primary cause
                                                                from excessive production, dysregulation, or inade-
   of RA and that different mechanisms may indepen-
                                                                quate inhibition that have gained most attention.1
   dently lead to synovial inflammation in susceptible in-
                                                                Very simply, cytokines can be classified into func-
   dividuals. It is important to recognize that, based on
                                                                tional groups based on the cells of origin, and all major
   current information, we do not propose that periodon-
                                                                types have been identified and located in inflamed
   tal pathogens cause, or are associated with, RA. The
                                                                synovial and periodontal tissues.
   main focus of our attention is directed not towards
                                                                   Periodontitis has very similar cytokine profiles to
   causality but rather associations between two chronic
                                                                RA,27,28 consisting of persistent high levels of proin-
   inflammatory conditions that may have common un-
                                                                flammatory cytokines, including IL-1b and tumor
   derlying pathogenic mechanisms.
                                                                necrosis factor-alpha (TNF-a), and low levels of cyto-
   Immunogenetics                                               kines which suppress the immunoinflammatory re-
      Periodontitis. It has been reported that more than        sponse such as IL-10 and transforming growth
   50% of the variance in several features of chronic peri-     factor-b (TGF-b). These cytokines, together with
   odontitis can be explained by genetic factors.19,20          low levels of tissue inhibitors of metalloproteinases
   Many of these interindividual variables relate to sever-     (TIMPs) and high levels of MMPs and prostaglandin E2
   ity of periodontal destruction, and other inflammatory        (PGE2), are associated with the active stages of peri-
   responses are attributed partly to the amount and type       odontitis.
   of cytokines that individuals produce.21 While the              The destruction of soft and hard tissues seen in RA is
   HLA-DR phenotype is not particularly strong for peri-        also the result of not only a large number of cytokines

J Periodontol • November 2005 (Suppl.)                                                           Bartold, Marshall, Haynes

but also the sustained presence of other effector mol-      RA. No differences were noted for the plaque and
ecules released by resident and migrating cells. To-        bleeding indices between the control and RA groups.
gether, these soluble mediators of inflammation are          The RA group did, however, have significantly more
able to induce degradation of collagen and proteogly-       missing teeth than the control group and a greater per-
cans either through direct or indirect means. Produc-       centage of these subjects had deeper pocketing com-
tion of the arachidonic acid metabolite PGE2 as well as     pared to the controls. The percentage of alveolar bone
the release of neutrophil-associated enzymes, such as       loss correlated positively with the principal parame-
neutrophil elastase and b-glucuronidase, together           ters of RA severity.
with the secretion of matrix metalloproteinases by             These two pilot studies have resulted in several sig-
macrophages and synoviocytes, all contribute signif-        nificant findings. Contrary to current dogma, RA pa-
icantly to the pathogenesis of RA.                          tients do not have impaired oral hygiene (judged by
                                                            plaque and bleeding scores). Perhaps more impor-
Formulation of the Hypothesis
                                                            tantly, it was noted that individuals with severe RA
On the basis of the above considerable similarities be-
                                                            are more likely to have advanced periodontitis and
tween the pathological and clinical features of RA and
                                                            vice versa. Although many RA patients take medica-
periodontitis (especially in the advanced and aggres-
                                                            tions that can reduce periodontal destruction (i.e.,
sive forms of these diseases), we have proposed that
                                                            NSAIDs and immunosuppressants), we have noted
in some susceptible individuals, there are common
                                                            significant periodontal destruction in these patients.
features of an underlying and presently unknown dys-
                                                            This indicates that prior to the development of RA
regulation of the inflammatory mechanism which pre-
                                                            symptoms, the periodontitis was most likely develop-
disposes these individuals to advanced, aggressive,
                                                            ing and not detected. Thus, disease duration may be
and severe forms of either disease.
                                                            a very important factor. Finally, in order to understand
Studies on Relationships Between Periodontitis              the interrelationships between periodontitis and RA, it
and Rheumatoid Arthritis                                    is necessary to categorize the disease on the basis of
To date, very few studies have examined the associa-        severity and duration (i.e., type of disease).
tion between RA and periodontal disease, and the               Recently, using an animal model, additional evi-
results have often been conflicting. For example,            dence has been presented to indicate a significant re-
Finnish studies found no correlation between peri-          lationship between periodontitis and rheumatoid
odontal disease and arthritis,29 while others3,30 sug-      arthritis.32 From this study it was reported that induc-
gest a higher prevalence of periodontal bone loss in        ing experimental arthritis in the rat (adjuvant arthritis)
RA. A major reason for these discrepancies relates          resulted in periodontal breakdown characterized
to the lack of uniformity in classifying the various        by alveolar bone loss and increased matrix metallo-
forms of periodontal disease and RA. Indeed most            proteinase activity in adjacent gingival tissues. Inter-
of the early studies,3,29-31 failed to take into account    estingly, all of these reactions occurred without
the various forms of RA and periodontal disease             manipulating the oral or subgingival microflora.
and, as a result, grouped all subjects as either having
RA or periodontal disease with little or no regard for      Osteoclast Activation and Vascular Damage – A
subclassification for more detailed analyses. In light       Common Pathway in Periodontitis and
of these limitations, it is clear there is a need to        Rheumatoid Arthritis?
re-examine the extent of the association between            Most recently, studies from our laboratory (unpub-
specific types of RA and periodontal disease. In par-        lished data) have begun to investigate the codistribu-
ticular, it is our thesis that the more aggressive or       tion of cytokines involved in vascular damage and
severe forms of periodontal disease and RA will show        bone resorption in biopsies from graded rheumatoid
a very close correlation in terms of coexistence.           arthritis and periodontitis lesions. Since the tumor ne-
   In our first pilot study,6 investigating self-reported    crosis factor (TNF)-like molecules and their receptors
disease experience, the prevalence of moderate to           have been shown to be involved in both processes, we
severe periodontitis was significantly elevated in in-       have chosen to study the receptor activator of nF-
dividuals with RA (unadjusted relative risk of 4.7).        kappa B ligand (RANKL), osteopretogerin (OPG),
In addition, the converse was also true in that peri-       and TNF-related apoptosis inducing ligand (TRAIL)
odontitis patients had a higher prevalence of RA com-       to determine at least one molecular mechanism com-
pared to the general population (unadjusted relative        mon to both conditions.
risk of 1.5).                                                   The cell surface TNF-like molecule, RANKL and its
   In a subsequent study, 65 patients attending a           receptor, RANK have been shown to be key factors
rheumatology clinic were studied for their level of peri-   regulating osteoclast formation and activation.33,34
odontitis and RA.2 A control group consisted of age-        It has been shown that when RANKL binds to RANK
and gender-matched individuals who did not have             on the surface of osteoclast precursors, these cells

Periodontitis and Rheumatoid Arthritis                                                        Volume 76 • Number 11 (Suppl.)

   differentiate to form mature osteoclasts. It is now clear      In response to proinflammatory cytokines TNF-a
   that RANKL, together with macrophage-colony stim-           and IL-1b, OPG mRNA expression was dramatically
   ulating factor (M-CSF), is required for osteoclast for-     enhanced, resulting in secretion of newly synthesized
   mation. The soluble TNF ‘‘receptor-like’’ molecule,         OPG and a reduction in cell-associated OPG. Such
   OPG, is a natural inhibitor of RANKL.35 OPG binds           findings are consistent with our observations in vivo
   to RANKL and prevents its ligation to RANK. The im-         for active RA and periodontitis lesions. Vascular dam-
   portance of these molecules in regulating bone me-          age due to apoptosis is thought to precede vascular
   tabolism has been demonstrated by transgenic and            calcification51 and contribute to atherosclerosis.52
   gene knock-out studies in mice.36 Since these factors       In addition, diabetic endothelial cell dysfunction is as-
   control physiologic osteoclast formation, it is reason-     sociated with DNA damage induced by poly (ADP-
   able to propose that they may also be key regulators        ribose) polymerase activation. The exact cause of
   of pathological bone resorption.37,38 Although              endothelial cell dysfunction is not known but it is pos-
   RANKL is normally provided by osteoblast-like cells         sible that molecules such as TRAIL, expressed in
   in bone,38,39 there are reports suggesting that lym-        nearby cells and tissues, may be important.38,53 Our
   phocytes present in rheumatoid tissues may be the           recent binding studies confirm that OPG binds to
   main source of RANKL in inflammatory arthritis.33,40         TRAIL, although with less affinity than RANKL, in vi-
   Furthermore, CD3+ T cells from the human rheuma-            tro, and blocks its activity (unpublished data). The fi-
   toid joint express RANKL and can promote osteoclast         nal piece of compelling evidence for the role of OPG in
   formation from rodent spleen precursors.40 In addi-         vascular damage comes from the fact that OPG
   tion to lymphocyte production of RANKL, inhibition          knock-out mice develop vascular calcification. It is
   of RANKL by OPG treatment in vivo reduces both              significant to note that calcification cannot be re-
   bone and cartilage destruction in a model of adjuvant       versed by systemic treatment with recombinant OPG
   arthritis.41                                                postpartum.50 This supports our concept that OPG
      Under certain conditions, human osteoclasts are          must be expressed within the endothelial cells, either
   derived from osteoclast precursor cells present in or       in an appropriate form or associated with other mole-
   near to the tissues of arthritic joints.42,43 More recent   cules, and this only occurs following normal synthesis
   reports in humans44,45 and animals41 show that              within the healthy endothelial cells.
   RANK/RANKL interactions may be required for oste-              In light of the above, we propose that at least one
   oclast formation and bone resorption in the RA joint.       underlying common molecular pathway in common
   Accordingly, we have recently demonstrated that OPG         between rheumatoid arthritis and periodontitis may
   and RANKL are expressed in biopsies of inflamed              lie within the RANK/OPG/TRAIL axis whereby OPG
   rheumatoid synovium and periodontitis lesions.46 In         decreases leading to decreased vascular protection.
   addition, we have found (unpublished data) that an-         In addition, with an increase in RANKL and TRAIL
   other ligand for OPG, TRAIL, is expressed in the both       within the tissues, not only is vascular damage pos-
   types of tissue (although not from the same patient).       sible, but significant activation of osteoclasts may
   In these studies we have noted that OPG decreases           result. This proposal still awaits verification.
   with inflammation, RANKL increases with inflamma-
   tion, and TRAIL increases with inflammation. These           COMMON PATHOGENESIS – COMMON
   findings may be of considerable significance in light         TREATMENT?
   of OPG’s ability to block the activity of TRAIL (and        Current and Emerging Therapies
   vice versa) and TRAIL’s anti-inflammatory proper-            Currently, the mainstream ‘‘first-line’’ modes of treat-
   ties.47                                                     ment for RA remain the NSAIDs such as aspirin, nap-
      The production of OPG by endothelial cells may be        roxen, diclofenac, and ibuprofen. Their mechanism of
   significant for reasons other than its effects on bone       action through the inhibition of cyclooxygenase
   metabolism, and there is now evidence to suggest that       (COX) synthesis produces both analgesic and antipy-
   OPG might also regulate endothelial cell function.          retic properties. While these medications are effective
   OPG has been reported to be required for endothelial        in reducing the pain symptoms in RA, they do not sig-
   cell survival and growth.48 In addition, OPG knock-out      nificantly alter its course.54
   mice have been shown to develop arterial calcifica-             The use of NSAIDs for management of periodontal
   tion49,50 as well as severe osteoporosis, suggesting        disease has been studied over the past 20 years.55-57
   that vascular endothelial expression of OPG may             While the results appear promising, the widespread
   have a role in vascular homeostasis.41 One of the           clinical use of these medications to alter the course
   most unexpected findings from our recent studies of          of periodontitis has not been universal. One particular
   diseased periodontal and synovial tissues was the ob-       problem with their use for the management of peri-
   servation that endothelial cells produce large amounts      odontitis appears to be a ‘‘rebound’’ effect to baseline
   of OPG (unpublished observations).                          following cessation of the medication.58

J Periodontol • November 2005 (Suppl.)                                                            Bartold, Marshall, Haynes

   With the discovery of two COX enzymes responsi-          of elevated IL-1 in inflamed tissues.72-74 Similarly,
ble for PGE2 production, designated COX-1 (constitu-        other studies have shown that blocking the activity
tively expressed) and COX-2 (inducible), a variety of       of another important inflammatory cytokine, TNF-a,
COX-2 inhibitors have been studied for their potential      has therapeutic efficacy in RA patients.75-78 The roles
to stop or slow down bone resorption. One of the first       of IL-1 and TNF antagonists in a primate model of
COX-2 inhibitors developed, tenidap, has been shown         periodontitis have demonstrated a reduction in the
to inhibit not only cyclooxygenase and PGE2 produc-         inflammatory infiltrate in close proximity to bone as
tion but also IL-1, IL-6, and TNF-a production. To          well as reduction in the formation of osteoclasts and
date, COX-2 inhibitors have not been thoroughly             reduced bone loss.79
studied for their potential to modify bone resorption          Clearly, many of these biologic agents, which tar-
in periodontitis.                                           get specific molecular events associated with acute
   In contrast to the NSAIDS, which do not signifi-          and chronic inflammation, have significant potential
cantly alter the course of RA, a newer family of med-       to alter clinical outcomes for both RA and periodontal
ications designated disease-modifying anti-rheumatic        disease. With the emerging understanding that RA
drugs (DMARDs) has been developed. To be classified          and periodontitis are multifactorial diseases, combi-
as a DMARD, the medication must demonstrate an              nation therapies that target multiple disease out-
ability to change the course of RA for at least 1 year      comes are also emerging. For example, in an
as evidenced by sustained improvement in function,          animal study, it was reported that the administration
decreased synovitis, and prevention of further joint        of a combination of a chemically modified tetracycline
damage.59 Examples of these medications include             (CMT-1) plus an NSAID, such as flurbiprofen or teni-
parenteral gold salts, methotrexate, sulfasalazine, hy-     dap, synergistically inhibited severe bone destruction
droxychloroquine (antimalarial drug), penicillamine,        in arthritic rats, with the suppression of MMP activity in
azathioprine, and leflunomide. A major drawback in           the joints.80,81 Similar encouraging results have been
the use of DMARDs is their considerable toxicity.60,61      reported for periodontitis in humans.82
   The use of DMARDs for the management of peri-               Notwithstanding the above, it must be recognized
odontitis has been restricted largely due to the toxicity   that periodontitis differs in one significant way from
issues. However, the use of gold salts in an animal         RA through our understanding that the subgingival
model has shown reduced periodontal destruction.62          biofilm is a key etiologic factor. Unlike periodontal
To date, no human studies have been performed.              disease, no specific bacterial etiology has been iden-
   Another emerging area of potential for host modu-        tified for RA. Thus, while host modification of disease
lation in periodontitis and rheumatoid arthritis is con-    processes are possible for periodontitis, controlling
trol of the MMPs that are important mediators of            the bacteria that cause periodontal infections remains
connective tissue breakdown in both hard and soft tis-      a significant focus for periodontal treatment and pre-
sues. In this regard, tetracyclines and various chemi-      vention. At best, host modification can be only an ad-
cal analogues have been found to inhibit MMP activity       junct treatment for periodontitis. However, until an
by a mechanism that is independent of their antimi-         etiologic factor can be found for RA, host modification
crobial property.63,64 A number of clinical trials using    remains the mainstay of treatment.
low-dose tetracycline to modify periodontitis have
been carried out, with the most recent data indicating      CONCLUSIONS
that low-dose doxycycline is safe and significantly ef-      There is no question that periodontitis and RA have
fective.65,66 Nonetheless, it is still recommended that     many pathologic features in common. Emerging evi-
these data be interpreted with caution to differentiate     dence suggests a strong relationship between the ex-
between statistically significant and clinically relevant    tent and severity of periodontal disease and RA. While
findings.67 The role of MMP inhibitors in managing RA        this relationship is unlikely to be causal, it is clear that
has been less well studied but promising results are        individuals with advanced RA are more likely to expe-
emerging.68-70 In particular, a recent study has dem-       rience more significant periodontal problems com-
onstrated that low-dose and antimicrobial (higher)          pared to their non-RA counterparts, and vice versa.
dose doxycycline, when used adjunctively with meth-         Hence, the possibility exists that both conditions re-
otrexate, produces enhanced improvements in global          sult from a common underlying dysregulation of the
scores of RA severity in humans than methotrexate           host inflammatory response. The precise nature of
combined with placebo.71                                    this dysregulation remains to be established.
   Control of cytokines and their receptors is also            There is accruing evidence to support the notion
emerging as a field of considerable promise. For ex-         that both conditions manifest as a result of an imbal-
ample, blocking the IL-1 receptor and using gene            ance between proinflammatory and anti-inflammatory
therapy to deliver IL-1 receptor antagonist are two         cytokines. As a result, new treatment strategies will
strategies under investigation to modulate the effect       emerge for both diseases that may target the inhibition

Periodontitis and Rheumatoid Arthritis                                                             Volume 76 • Number 11 (Suppl.)

   of proinflammatory cytokines and destructive pro-              16. Page RC, Offenbacher S, Schroeder HE, Seymour GJ,
   teases.                                                           Kornman KS. Advances in the pathogenesis of peri-
                                                                     odontitis: Summary of developments, clinical implica-
      Through a better understanding of these two com-
                                                                     tions and future directions. Periodontol 2000 1997;
   mon chronic inflammatory conditions, it is hoped that              14:216-248.
   areas of similarity can be exploited to determine the         17. Ebringer A, Wilson C. HLA molecules, bacteria and
   true relationship between these diseases and com-                 autoimmunity. J Med Microbiol 2000;49:305-310.
   mon areas of treatment. Already, it can be predicted          18. Zambon JJ. Periodontal diseases: Microbial factors.
                                                                     Ann Periodontol 1996;1:879-925.
   that the periodontal status of patients with RA should
                                                                 19. Michalowicz BS, Aeppli DP, Kuba RK, et al. A twin
   be carefully screened.                                            study of genetic variation in proportional radiographic
                                                                     alveolar bone height. J Dent Res 1991;70:1431-
   REFERENCES                                                        1435.
    1. Greenwald RA, Kirkwood K. Adult periodontitis as          20. Hassell T, Harris E. Genetic influences in caries and
       a model for rheumatoid arthritis with emphasis on             periodontal diseases. Crit Rev Oral Biol Med 1995;
       treatment strategies. J Rheumatol 1999;26:1650-               6:319-342.
       1653.                                                     21. Salvi GE, Brown CE, Fujihashi K, et al. Inflammatory
    2. Mercado FB, Marshall RI, Klestov AC, Bartold PM.              mediators of the terminal dentition in adult and early
       Relationship between rheumatoid arthritis and peri-           onset periodontitis. J Periodontal Res 1998;33:212-
       odontitis. J Periodontol 2001;72:779-787.                     225.
    3. Kasser UR, Gleissner C, Dehne F, Michael A, Willer-       22. Michalowicz BS, Diehl SR, Gunsolley JC, et al. Evi-
       shausen-Zonuchen B, Bolten WW. Risk for periodontal           dence of a substantial genetic basis for adult peri-
       disease in patients with longstanding rheumatoid              odontitis. J Periodontol 2000;71:1699-1707.
       arthritis. Arthritis Rheum 1997;40:2248-2251.             23. Kornman KS, Crane A, Wang H-Y, et al. The in-
    4. Gleissner C, Willershausen B, Kaesser U, Bolten WW.           terleukin-1 genotype as a severity factor in adult
       The role of risk factors for periodontal disease in           periodontal disease. J Clin Periodontol 1997;24:72-
       patients with rheumatoid arthritis. Eur J Med Res             77.
       1998;3:387-392.                                           24. Kurosaka M, Ziff M. Immunoelectron microscopic
    5. Albandar JM. Some predictors of radiographic alveo-           study of the distribution of T cell subsets in rheumatoid
       lar bone height reduction over 6 years. J Periodontal         synovium. J Exp Med 1983;158:1191-1210.
       Res 1990;25:186-192.                                      25. Nepom GT, Nepom BS. Prediction of suscepti-
    6. Mercado F, Marshall RI, Klestov AC, Bartold PM. Is            bility to rheumatoid arthritis by human leukocyte
       there a relationship between rheumatoid arthritis and         antigen genotyping. Rheum Dis Clin Am 1992;18:
       periodontal disease? J Clin Periodontol 2000;127:267-         785-792.
       272.                                                      26. Gregersten PK, Silver J, Winchester R. The shared
    7. DeStefano F, Anda RF, Kahn HS, Williamson DF,                 epitope hypothesis. An approach to understanding the
       Russell CM. Dental disease and risk of coronary heart         molecular genetics of susceptibility to rheumatoid
       disease and mortality. Br Med J 1993;306:688-691.             arthritis. Arthritis Rheum 1987;30:1205-1213.
    8. Beck J, Garcia R, Heiss G, Vokonas PS, Offenbacher S.     27. Snyderman R, McCarty GA. Analogous mechanisms
       Periodontal disease and cardiovascular disease. J             of tissue destruction in rheumatoid arthritis and peri-
       Periodontol 1996;67:1123-1137.                                odontal disease. In: Genco RJ, Mergenhagen SE, eds.
    9. Yalda B, Offenbacher S, Collins JG. Diabetes as a             Host-Parasite Interaction in Periodontal Diseases.
       modifier of periodontal disease expression. Periodontol        Washington, DC: American Society for Microbiology;
       2000 1994;6:37-49.                                            1982:354-362.
   10. Weyand CM. New insights into the pathogenesis of          28. Arend WP, Dayer JM. Cytokines and cytokine inhib-
       rheumatoid arthritis. Rheumatology 2000;39(Suppl. 1):         itors or antagonists in rheumatoid arthritis. Arthritis
       3-8.                                                          Rheum 1990;33:305-315.
   11. Hirschfeld L, Wasserman B. A long-term survey of          29. Helminen-Pakkala E. Periodontal conditions in rheu-
       tooth loss in 600 treated periodontal patients. J Peri-       matoid arthritis. A clinical and roentenological inves-
       odontol 1978;49:225-237.                                      tigation. Part two. The study in rheumatoids. Proc
   12. O’Sullivan JB, Cathcart ES. The prevalence of rheu-           Finnish Dental Soc 1971;Suppl IV:1-108.
       matoid arthritis. Follow-up evaluation of the effect of   30. Tolo K, Jorkjend L. Serum antibodies and loss of peri-
       criteria on rates in Sudbury, Massachusetts. Ann              odontal bone in patients with rheumatoid arthritis. J Clin
       Intern Med 1972;76:573-577.                                   Periodontol 1990;17:288-291.
   13. Pincus T, Marcum SB, Callahan LF. Long-term drug          31. Risheim H, Kjaerheim V, Arneberg P. Improvement of
       therapy for rheumatoid arthritis in seven rheumatology        oral hygiene in patients with rheumatoid arthritis.
       private practices: II. Second line drugs and predni-          Scand J Dent Res 1992;100:172-175.
       sone. J Rheumatol 1992;19:1885-1894.                      32. Ramamurthy NS, Greenwald RA, Celiker MY, Shi EY.
   14. Pincus T, Marcum SB, Callahan LF, et al. Long-term            Experimental arthritis in rats induces biomarkers of
       drug therapy for rheumatoid arthritis in seven rheu-          periodontitis which are ameliorated by gene therapy
       matology private practices. I. Non-steroidal anti-in-         with tissue inhibitor of matrix metalloproteinases. J
       flammatory drugs. J Rheumatol 1992;19:1874-                    Periodontol 2005;76:229-233.
       1884.                                                     33. Yasuda H, Shima N, Nakagawa N, et al. Osteoclast
   15. Wolfe F, Hawley DJ, Cathey MA. Termination of slow            differentiation factor is a ligand for osteoprotegerin/
       acting antirheumatic therapy in rhematoid arthritis: A        osteoclastogenesis-inhibitory factor and is identical to
       14-year prospective evaluation of 1017 consecutive            TRANCE/RANKL. Proc Natl Acad Sci USA 1998;95:
       starts. J Rheumatol 1990;17:994-1002.                         3597-3602.

J Periodontol • November 2005 (Suppl.)                                                               Bartold, Marshall, Haynes

34. Lacey DL, Timms E, Tan HL, et al. Osteoprotegerin          51. Proudfoot D, Skepper JN, Hegyi L, Farzaneh-Far A,
    ligand is a cytokine that regulates osteoclast differen-       Shanahan CM, Weissberg PL. The role of apoptosis in
    tiation and activation. Cell 1998;93:165-176.                  the initiation of vascular calcification. Z Kardiol
35. Simonet WS, Lacey DL, Dunstan CR, et al. Osteopro-             2000;90S3:43-46.
    tegerin: A novel secreted protein involved in the          52. Dimmeler S, Hermann C, Zeiher AM. Apoptosis of
    regulation of bone density. Cell 1997;89:309-319.              endothelial cells. Contribution to the pathophysiology
36. Abu-Amer Y, Abbas S, Hirayama T. TNF receptor type             of atherosclerosis. Eur Cytokine Netw 1998;9:697-
    1 regulates RANK ligand expression by stromal cells            698.
    and modulates osteoclastogenesis. J Cell Biochem           53. Gochuico BR, Zhang J, Ma BY, Marshak-Rothstein A,
    2004;93:980-989.                                               Fine A. TRAIL expression in vascular smooth muscle.
37. Haynes DR, Crotti TN, Potter AE, et al. The osteoclas-         Am J Physiol Lung Cell Mol Physiol 2000;278:L1045-
    togenic modules RANKL and RANK are associated                  L1050.
    with periprosthetic osteolysis. J Bone Joint Surg Br       54. Lipsky PE. Rheumatoid arthritis. In: Wilson JD, Braun-
    2001;83:902-911.                                               wald E, Isselbacher KJ, et al., eds. Harrison’s Princi-
38. Haynes DR, Crotti TN, Loric M, Bain GI, Atkins GJ,             ples of Internal Medicine, 12th ed. New York: McGraw-
    Findlay DM. Osteoprotegerin and receptor activator of          Hill; 1991:1437-1443.
    nuclear factor kappaB ligand (RANKL) regulate oste-        55. Feldman RS, Szeto B, Chauncey HH, Goldhaber P.
    oclast formation by cells in the human rheumatoid              Non-steroidal anti-inflammatory drugs in the reduc-
    arthritic joint. Rheumatology 2001;40:623-630.                 tion of human alveolar bone loss. J Clin Periodontol
39. Teng Y-TA, Nguyen H, Gao X, et al. Functional human            1983;10:131-136.
    T-cell immunity and osteoprotegerin ligand control         56. Jeffcoat MK, Page R, Reddy M, et al. Use of digi-
    alveolar bone destruction in periodontal infection. J          tal radiography to demonstrate the potential of nap-
    Clin Invest 2000;106:R59-R67.                                  roxen as an adjunct in the treatment of rapidly
40. Horwood NJ, Kartsogiannis V, Quinn JMW, Romas E,               progressive periodontitis. J Periodontal Res 1991;26:
    Martin TJ, Gillespie MT. Activated T lymphocytes               415-421.
    support osteoclast formation in vitro. Biochem Bio-        57. Paquette DW, Williams RC. Modulation of host in-
    phys Res Commun 1999;264:144-150.                              flammatory mediators as a treatment strategy for
41. Kong Y-Y, Feige U, Sarosi I, et al. Activated T cells          periodontal diseases. Periodontol 2000 2000;24:239-
    regulate bone loss and joint destruction in adjuvant           252.
    arthritis through osteoprotegerin ligand. Nature           58. Williams RC, Jeffcoat MK, Howell TH, et al. Altering
    1999;402:304-309.                                              the progression of human alveolar bone loss with the
42. Fujikawa Y, Sabokbar A, Neale S, Athanasou NA.                 non-steroidal anti-inflammatory drug flurbiprofen. J
    Human osteoclast formation and bone resorption by              Periodontol 1989;60:485-490.
    monocytes and synovial macrophages in rheumatoid           59. Paget S. Treatment. In: Klippel J, ed. Primer on the
    arthritis. Ann Rheum Dis 1996;55:816-822.                      Rheumatic Diseases. Atlanta: Arthritis Foundation;
43. Toritsuka Y, Nakamura N, Lee SB, et al. Osteoclasto-           1997:168-174.
    genesis in iliac bone marrow of patients with rheuma-      60. American College of Rheumatology Ad Hoc Commit-
    toid arthritis. J Rheumatol 1997;24:1690-1696.                 tee on Clinical Guidelines. Guidelines for monitoring
44. Gravallese EM, Manning C, Tsay A, et al. Synovial              drug therapy in the management of rheumatoid
    tissue in rheumatoid arthritis is a source of osteoclast       arthritis. Arthritis Rheum 1996;39:713-722.
    differentiation factor. Arthritis Rheum 2000;43:250-       61. American College of Rheumatology Ad Hoc Commit-
    258.                                                           tee on Clinical Guidelines. Guidelines for monitoring
45. Takayanagi H, Iizuka H, Juji T, et al. Involvement of          drug therapy in the management of rheumatoid
    receptor activator of nuclear factor k ligand/osteoclast       arthritis. Arthritis Rheum 1996;39:723-731.
    differentiation factor in osteoclastogenesis from syno-    62. Novak MJ, Polson AM, Freeman E. Effects of gold salts
    viocytes in rheumatoid arthritis. Arthritis Rheum              on experimental periodontitis. I. Histometric evalua-
    2000;43:259-269.                                               tion of periodontal destruction. J Periodontol 1984;
46. Crotti T, Smith MD, Hirsch R, et al. Receptor activator        55:69-77.
    NF kB ligand (RANKL) and osteoprotegerin (OPG)             63. Golub LM, Lee HM, Lehrer G, et al. Minocycline
    protein expression in periodontitis. J Periodontal Res         reduces gingival collagenolytic activity during diabe-
    2003;38:380-387.                                               tes: Preliminary observations and a proposed new
47. Song K, Chen Y, Goke R, et al. Tumor necrosis factor-          mechanism of action. J Periodontal Res 1983;
    related apoptosis-inducing ligand (TRAIL) is an in-            18:516-526.
    hibitor of autoimmune inflammation and cell cycle           64. Golub LM, McNamara TF, D’Angelo G, Greenwald RA,
    progression. J Exp Med 2000;191:1095-1134.                     Ramamurthy NS. A non-antibacterial chemically
48. Malyankar UM, Scatena M, Suchland KL, Yun TJ,                  modified tetracycline inhibits mammalian collagenase
    Clark EA, Giachelli CM. Osteoprotegerin is an avb3-            activity. J Dent Res 1987;66:1310-1314.
    induced, NF-kB-dependent survival factor for endo-         65. Reddy MS, Geurs NC, Gunsolley JC. Periodontal host
    thelial cells. J Biol Chem 2000;275:20959-20962.               modulation with antiproteinase, anti-inflammatory,
49. Bucay N, Sarosi I, Dunstan CR, et al. Osteoprotegerin-         and bone-sparing agents. A systematic review. Ann
    deficient mice develop early onset osteoporosis and             Periodontol 2003;8:12-37.
    arterial calcification. Genes Dev 1998;12:1260-1268.        66. Preshaw PM, Hefti AF, Novak MJ, et al. Subantimicro-
50. Min H, Morony S, Sarosi I, et al. Osteoprotegerin              bial dose doxycycline enhances the efficacy of scaling
    reverses osteoporosis by inhibiting endosteal osteo-           and root planing in chronic periodontitis: A multicenter
    clasts and prevents vascular calcification by blocking          trial. J Periodontol 2004;75:1068-1076.
    a process resembling osteoclastogenesis. J Exp Med         67. Greenstein G. Efficacy of subantimicrobial-dose doxy-
    2000;192:463-474.                                              cycline in the treatment of periodontal diseases: A

Periodontitis and Rheumatoid Arthritis                                                                 Volume 76 • Number 11 (Suppl.)

         critical evaluation. Int J Periodontics Restorative Dent          arthritis receiving methotrexate. N Engl J Med 1999;
         2004;24:528-543.                                                  340:253-259.
   68.   O’Dell JR, Haire CE, Palmer W, et al. Treatment             77.   Moreland LW, Margolies G, Heck LW Jr, et al.
         of early rheumatoid arthritis with minocycline or                 Recombinant soluble tumor necrosis factor receptor
         placebo: Results of a randomized, double-blind,                   (p80) fusion protein: Toxicity and dose finding trial in
         placebo-controlled trial. Arthritis Rheum 1997;40:                refractory rheumatoid arthritis. J Rheumatol
         842-848                                                           1996;23:1849-1855.
   69.                     ´
         Tilley BC, Alarcon GS, Heyse SP, et al. Minocycline in      78.   Elliot MJ, Maini RN, Feldmann M, et al. Randomised
         rheumatoid arthritis: A 48-week, double-blind pla-                double-blind comparison of chimeric monoclonal an-
         cebo-controlled trial. Ann Intern Med 1995;122:81-                tibody to tumor necrosis factor a (cA2) versus placebo
         89.                                                               in rheumatoid arthritis. Lancet 1994;344:1105-1110.
   70.   Voils SA, Evans ME, Lane MT, Schosser RH, Rapp RP.          79.   Assuma R, Oates T, Cohran D, Amar S, Graves DT. IL-
         Use of macrolides and tetracyclines for chronic in-               1 and TNF antagonists inhibit the inflammatory re-
         flammatory diseases. Ann Pharmacother 2005;39:86-                  sponse and bone loss in experimental periodontitis. J
         94.                                                               Immunol 1998;160:403-409.
   71.   Elliott JR, Paulsen GA, Mallek JA, et al. Treatment of      80.   Greenwald RA, Moak SA, Ramamurthy NS, Golub LM.
         early sero-positive rheumatoid arthritis: Doxycycline             Tetracyclines suppress matrix metalloproteinase ac-
         with methotrexate versus methotrexate alone. Arthritis            tivity in adjuvant arthritis and, in combination with
         Rheum 2003;48(Suppl.):S654.                                       flurbiprofen, ameliorate bone damage. J Rheumatol
   72.   Roessler BJ, Allen ED, Wilson JM, Hartman JW,                     1992;19:927-938.
         Davidson BL. Adenoviral-mediated gene transfer to           81.   Leung MK, Greenwald RA, Ramamurthy NS, et al.
         rabbit synovium in vivo. J Clin Invest 1993;92:1085-              Tenidap and flurbiprofen enhance uptake of matrix
         1092.                                                             metalloproteinase inhibitor 4-dedimethylamino tetra-
   73.   Bandara G, Mueller GM, Galea-Lauri J, et al. Intra-               cycline in inflamed joints of adjuvant arthritic rats. J
         articular expression of biologically active interleukin-1         Rheumatol 1995;22:1726-1731.
         receptor antagonist protein by ex vivo gene transfer.       82.   Lee HM, Ciancio SG, Tuter G, Ryan ME, Komaroff E,
         Proc Natl Acad Sci USA 1993;90:10764-10768.                       Golub LM. Subantimicrobial dose doxycycline efficacy
   74.   Evans CH, Ghivizzani SC, Kang R, et al. Gene therapy              as a matrix metalloproteinase inhibitor in chronic
         for rheumatoid diseases. Arthritis Rheum 1999;42:1-               periodontitis patients is enhanced when combined
         16.                                                               with a non-steroidal anti-inflammatory drug. J Peri-
   75.   Maini RN, Breedveld FC, Kalden JR, et al. Therapeutic             odontol 2004;75:453-463.
         efficacy of multiple intravenous infusions of anti-tumor
         necrosis factor a monoclonal antibody combined with         Correspondence: Dr. P.M. Bartold, University of Adelaide,
         low-dose weekly methotrexate in rheumatoid arthritis.       Department of Dentistry, Frome Road, Adelaide, South
         Arthritis Rheum 1998;41:1552-1563.                          Australia 5005, Australia. Fax: 61-8-8303-3435; e-mail:
   76.   Weinblatt ME, Kremer JM, Bankhurst AD, et al. A trial
         of etanercept, a recombinant tumor necrosis factor
         receptor:Fc fusion protein, in patients with rheumatoid     Accepted for publication March 7, 2005.