J Periodontol 2013;84(4 Suppl.):S20-S23doi:10.1902/jop.2013.1340020

Periodontitis and systemicdiseases: a record of discussionsof working group 4 of the JointEFP/AAP Workshop onPeriodontitis and SystemicDiseasesLinden GJ, Herzberg MC and on behalf of working group 4 of the joint EFP/AAPworkshop. Periodontitis and systemic diseases: a record of discussions of workinggroup 4 of the Joint EFP/AAP Workshop on Periodontitis and Systemic Diseases.

AbstractBackground: There has been an explosion in research into possible associationsbetween periodontitis and various systemic diseases and conditions.Aim: To review the evidence for associations between periodontitis and various sys-temic diseases and conditions, including chronic obstructive pulmonary disease(COPD), pneumonia, chronic kidney disease, rheumatoid arthritis, cognitive impair-ment, obesity, metabolic syndrome and cancer, and to document headline discus-sions of the state of each field. Periodontal associations with diabetes, cardiovasculardisease and adverse pregnancy outcomes were not discussed by working group 4.Results: Working group 4 recognized that the studies performed to date were lar-gely cross-sectional or case-control with few prospective cohort studies and norandomized clinical trials. The best current evidence suggests that periodontitis ischaracterized by both infection and pro-inflammatory events, which variouslymanifest within the systemic diseases and disorders discussed. Diseases with atleast minimal evidence of an association with periodontitis include COPD, pneu-monia, chronic kidney disease, rheumatoid arthritis, cognitive impairment, obes-ity, metabolic syndrome and cancer. The working group agreed that there isinsufficient evidence to date to infer causal relationships with the exception thatorganisms originating in the oral microbiome can cause lung infections.Conclusions: The group was unanimous in their opinion that the reported associ-ations do not imply causality, and establishment of causality will require new studiesthat fulfil the Bradford Hill or equivalent criteria. Precise and community-agreedcase definitions of periodontal disease states must be implemented systematicallyto enable consistent and clearer interpretations of studies of the relationship tosystemic diseases. The members of the working group were unanimous in theiropinion that to develop data that best inform clinicians, investigators and thepublic, studies should focus on robust disease outcomes and avoid surrogate end-points. It was concluded that because of the relative immaturity of the body ofevidence for each of the purported relationships, the field is wide open and thegaps in knowledge are large.

Gerry J. Linden1, Mark C. Herzberg2

and on behalf of working group 4 ofthe joint EFP/AAP workshop*1Centre for Public Health, School of

Medicine, Dentistry and Biomedical Sciences,

Queen’s University Belfast, Belfast, Northern

Ireland, UK; 2Department of Diagnostic and

Biological Science, School of Dentistry,

University of Minnesota, Minneapolis, MN,

USA

Key words: cancer; chronic kidney disease;

cognitive impairment; epidemiology;

metabolic syndrome; obesity; periodontal

diseases; periodontitis; pneumonia;

respiratory disease; rheumatoid arthritis

Accepted for publication 14 November 2012

The proceedings of the workshop were jointly

and simultaneously published in the Journal

of Clinical Periodontology and Journal of

Periodontology.

© 2013 European Federation of Periodontology and American Academy of PeriodontologyS20

Introduction

This report outlines the discussionsof working group 4 who were taskedwith reviewing possible associationsbetween periodontitis and less com-mon systemic diseases and condi-tions. At the outset, there wasunanimity within the group that thebody of evidence for many of thepurported relationships was rela-tively immature, and therefore, thisreport does not represent a tradi-tional consensus view, more a recordof discussions and recommendationsto strengthen the evidence base inthe future.

Association Versus Causation

It was accepted that scientific studiesthat show an association between agiven factor and a health effect can-not be extrapolated to imply that thefactor causes the specific disease. Forexample, a large number of earlyepidemiological studies suggested

that women taking hormone replace-ment therapy (HRT) had a reducedincidence of coronary heart disease(CHD). From these data, one mightconclude that HRT is protectiveagainst CHD. Data from subsequentrandomized controlled trials, how-ever, support the conclusion thatHRT causes a significant increase inthe risk of CHD (Rossouw et al.2002). Therefore, a cautious evalua-tion of any association must beundertaken and conclusions shouldbe balanced in favour of the morerobust study designs. The group alsodiscussed the value of using the nineBradford Hill criteria (Hill 1965) toestablish the strength of evidence forcomplex conditions of infective aeti-ology, but where classical Koch’spostulates cannot be satisfied.Hence, the body of available evi-dence should be evaluated for thefollowing:

(a) Statistical strength of association(b) Consistency(c) Specificity

(d) Temporal relationship (e.g. causeprecedes consequence)

(e) Biological gradient or dose–response relationship (e.g. moreperiodontitis leads to more ath-erosclerosis)

(f) Biological plausibility(g) Coherence(h) Experimental reversibility(i) Analogy – other precedents

Definitions of Disease

Epidemiological studies often analyseand encourage conclusions based onsurrogate markers for example mea-sures of atherosclerosis rather thancardiovascular events such as myo-cardial infarction. The workinggroup agreed that data using surro-gate markers often do not explainactual disease events and studiesintended to impact upon humanhealth and provider behavioursshould strive to investigate healthoutcomes rather than intermediate orsurrogate markers of disease process.

*Working group participants:Mark Bartold, Australia; Michael Glick, USA; Mark Herzberg, USA; Francis Hughes, UK; Palle Holmstrup, Denmark; Gerry Lin-den, UK; Angelo Mariotti, USA; Marc Quirynen, Belgium; Frank Scannapieco, USA; Arie Jan van Winkelhoff, The Netherlands.

Conflict of interest and source of funding

Group participants declare no conflict of interest. The workshop was funded by an unrestricted educational grant from Colgate-Palmolive to the European Federation of Periodontology and the American Academy of Periodontology.

© 2013 European Federation of Periodontology and American Academy of Periodontology

Periodontitis and systemic diseases S21

The group unanimously agreedthat precise and community-agreedcase definitions of periodontal dis-ease status must be implemented sys-tematically to enable reasonableinterpretations of studies of theirrelationship to systemic diseases. Theworking group recognized that pur-ported associations are often unclearbecause periodontal disease is a het-erogeneous mix of conditions. Thisproblem may be further confoundedby imprecisely defined systemic out-comes (e.g. cardiovascular disease,which might include atherosclerosis,vasospasm, myocarditis, heartischaemia and myocardial infarction)in the target disease. Meaningfulrelationships between the diseaseoutcomes and measured endpointswill likely depend on a strict andnarrow definition of the diseasesunder study. It was accepted thatnarrower case definitions will increasethe likelihood of identifying a mean-ingful association should one exist.

Study Design

The working group recognized thatstudies of the relevant systemic dis-ease associations with periodontitisand oral health are in their infancy.For chronic obstructive pulmonarydisease (COPD), pneumonia, chronickidney disease, rheumatoid arthritis,cognitive impairment, obesity, meta-bolic syndrome and cancer, the studiesthat have been performed are largelycross-sectional, case-control studieswith few prospective cohort studies.The working group also recognizedan absence of periodontal interventionstudies or randomized clinical trials.

What Lesser Studied Diseases and

Conditions are Reported to beAssociated with Periodontitis and

Other Oral Diseases?

The working group discussed onlythose diseases and conditions thatwere reviewed and evaluated by Lin-den et al. (2013). Better-known rela-tionships with periodontitis such asdiabetes (Borgnakke et al. 2013),cardiovascular disease (Dietrich et al.2013) and adverse pregnancy out-comes (Ide & Papapanou 2013) werethe subject of other reviews andwere not discussed. As described inLinden et al. (2013), diseases andconditions that show at least mini-

mal evidence of an association withperiodontitis include COPD, pneu-monia, chronic kidney disease, rheu-matoid arthritis, cognitive impairment,obesity, metabolic syndrome andcancer. It was accepted that associa-tions can be unidirectional or bidirec-tional, however, the working groupgenerally focused on unidirectionalassociations whereby periodontitisand other oral conditions could influ-ence systemic diseases. Interestingly,some conditions such as obesity mayaffect periodontal outcomes. Thegroup also recognized that otherassociations with systemic conditionsnot considered here may exist.

Does the Diversity of SystemicMedical Conditions Associated with

Periodontitis and Other OralConditions Reflect a Common

Underlying Mechanism(s)?

The working group agreed that thebest evidence to date suggested thatinfection and subsequent pro-inflam-matory events occur in periodontitisand may variously contributetowards select systemic diseases anddisorders. No causal relationships canbe inferred to date, with the excep-tion that organisms found in the oralmicrobiome cause lung infections.

The working group discussedhow the systemic impact of a specificperiodontal condition may be con-founded by the subject’s infectiousdisease history. As is the case forperiodontal conditions, other infectionscan disseminate infectious and pro-inflammatory stimuli, which mayhave a chronic systemic impact. Thespecific contributions of periodontalinfections to systemic disorders,therefore, may be confounded bytotal pathogen burden over time(Kinane & Bouchard 2008).

Review of the Purported PeriodontalDisease Associations with Specific

Systemic Diseases and Conditions

The working group agreed the fol-lowing:

Chronic obstructive pulmonary

disease

An association with periodontitis issuggested based on analyses of theNHANES data sets. There appearsto be a dose effect, whereby greaterperiodontal disease is associated with

increasing loss of lung function. Theprimary aetiological factor is smok-ing as modified by underlyinginflammation. It is plausible that theinflammatory status may be modifiedeither by aspiration of dental plaqueand/or hematogenous disseminationof inflammatory mediators and pla-que organisms from periodontalpockets. Studies of the associationbetween periodontal disease andexacerbations of COPD would bevaluable.

Pneumonia

The association between dental pla-que and pneumonia appears to bestronger than for plaque and COPD.That improved oral hygiene reducesthe risk of health care-associatedpneumonia is suggested by severalmeta-analyses. The relationship withperiodontitis however is not known.

Chronic kidney disease

The association between chronic kid-ney disease (CKD) and periodontitisin several studies is statistically sig-nificant and consistent. Althoughhypertension and diabetes are theprimary aetiological factors, peri-odontitis is hypothesized to modifyboth these aetiological factors andconsequently the presentation ofCKD. Studies of the effects of peri-odontal interventions on the emer-gence of CKD in patients with type2 diabetes would be useful, particu-larly to adjust for confounding ofdiabetes with this condition. Similarstudies examining the effect of peri-odontal interventions on CKD pro-gression are warranted.

Rheumatoid arthritis

Reports of an epidemiological asso-ciation with periodontal disease,including the NHANES data set andcase-control studies, are inconsistent.Animal studies provide biologicalplausibility. For example, Porphyro-monas gingivalis can induce andexacerbate a rheumatoid arthritis-like condition in susceptible rodents(Kinloch et al. 2011). In humanrheumatoid arthritis, antibodiesagainst citrullinated proteins andpeptides are often detected in theblood reflecting the enzymic conver-sion of arginine residues to citrulline

© 2013 European Federation of Periodontology and American Academy of Periodontology

S22 Linden et al.

in certain proteins. Biological plausi-bility for an association with peri-odontitis is also reflected in thedetection of citrullinated proteins ininflamed gingiva, which may beassociated with elevated auto-anti-bodies to self-antigens (Nesse et al.2012).

Cognitive impairment

The results of epidemiological stud-ies are difficult to interpret. Datafrom existing studies reflect imprecisecase definitions for periodontitis andthe heterogeneity of cases of cog-nitive impairment. There are fewstudies that point to underlying bio-logical plausibility.

Obesity

Many clinical studies of obesity andperiodontal disease have beenreported including a number of sys-tematic reviews. There are a few pro-spective studies indicating anassociation with periodontitis. Morespecifically, the literature suggeststhat obesity might adversely affectperiodontitis, but there is little evi-dence from clinical data or for bio-logical plausibility that periodontitismay affect obesity.

Metabolic syndrome

Metabolic syndrome is defined as acomplex of five clinical signs (asdefined by the International DiabetesFederation), which include centralobesity, raised triglycerides, reducedHDL cholesterol, raised blood pres-sure and raised fasting blood glucose.Syndromic presentation of obesityand at least two other signs of meta-bolic syndrome indicate that thepatient meets the diagnostic criteria.Evidence in support of a relationshipwith periodontitis needs to be morerobustly developed. Relationships

with one or two of these clinical signshave been reported but whether thesefindings point to an association withmetabolic syndrome is questionable.The limitations of the existing dataare discussed further in the review byLinden et al. (2013).

Cancer

Prospective and case-control studiessuggest an association between peri-odontitis and oral and oro-pharyn-geal cancer. Whereas some oralmicrobes can alter cells and tissuesconsistent with features of malignantchanges, additional biological plausi-bility, clinical and robust epidemio-logical data are needed to strengthensuch associations.

Conclusions

Working group 4 agreed that theassociations reviewed do not implycausality which will require newstudies and fulfilment of the Brad-ford Hill or equivalent criteria. Sincethe systemic diseases under consider-ation are complex and often hetero-geneous, it was accepted that there isgenerally justification to classify andanalyse subgroups. The workinggroup concluded that to have great-est public health utility, studiesshould focus on disease outcomesand avoid the use of biomarkers andsurrogate measures. It was agreedthat to assist the community in eval-uating the quality of evidence, allclinical trials must be registered withinternationally compliant registriesas appropriate to the locale. Work-ing group 4 accepted that the rela-tive immaturity of the body ofevidence for each of the purportedrelationships meant the field is wideopen and the gaps in knowledgeremain large.

References

Borgnakke, W., Ylostalo, P., Taylor, G. & Gen-co, R. (2013) Effect of periodontal disease ondiabetes: systematic review of epidemiologicobservational evidence. Journal of Clinical Peri-odontology 40 (Suppl 14), 135–152.

Dietrich, T., Sharma, P., Walter, C., Weston, P.& Beck, J. (2013) The epidemiological evidencebehind the association between periodontitisand incident atherosclerotic cardiovascular dis-ease. Journal of Clinical Periodontology 40(Suppl 14), 70–84.

Hill, A. B. (1965) Environment and disease: asso-ciation or causation. Proceedings of the RoyalSociety of Medicine 58, 295–300.

Ide, M. & Papapanou, P. N. (2013) Epidemiologyof association between maternal periodontaldisease and adverse pregnancy outcomes-sys-tematic review. Journal of Clinical Periodontol-ogy 40 (Suppl 14), 181–194.

Kinane, D. & Bouchard, P. (2008) Periodontaldiseases and health: consensus report of thesixth European workshop on periodontology.Journal of Clinical Periodontology 35, 333–337.

Kinloch, A. J., Alzabin, S., Brintnell, W., Wilson,E., Barra, L., Wegner, N., Bell, D. A., Cairns,E. & Venables, P. J. (2011) Immunization withPorphyromonas gingivalis enolase induces.Arthritis and Rheumatism 63, 3818–3823.

Linden, G. J., Lyons, A. & Scannapieco, F.(2013) Periodontal systemic associations.Review of the evidence. Journal of ClinicalPeriodontology 40 (Suppl 14), 8–19.

Nesse, W., Westra, J., van der Wal, J., Abbas, F.,Nicholas, A., Vissink, A. & Brouwer, E. (2012)The periodontium of periodontitis patients con-tains citrullinated proteins which may play arole in ACPA (anti-citrullinated protein anti-body) formation. Journal of Clinical Periodon-tology 39, 599–607.

Rossouw, J. E., Anderson, G. l., Prentice, R. L.,LaCroix, A. Z., Kooperberg, C., Stefanick, M.L., Jackson, R. D., Beresford, S., Howard, B.V., Johnston, K. C., Kotchen, M. & Ockene, J.(2002) Risks and benefits of estrogen plus pro-gestin in healthy postmenopausal women-prin-cipal results from the women’s health initiativerandomized controlled trial. Journal of theAmerican Medical Association 288, 321–333.

Address:Gerry LindenPeriodontal DepartmentSchool of DentistryQueen’s University,Grosvenor RoadBelfast BT12 6BPUKE-mail: g.linden@qub.ac.uk

Clinical Relevance

Many studies have identified asso-ciations between periodontal dis-ease and systemic diseases and

conditions. Patients with periodonti-tis are increasingly aware of researchinto possible links with systemic dis-eases. Dentists should be aware that

while outcomes to date supportweak associations there is no evi-dence that periodontal disease hasa causative role.

© 2013 European Federation of Periodontology and American Academy of Periodontology

Periodontitis and systemic diseases S23

Periodontitis and atheroscleroticcardiovascular disease:consensus report of the Joint EFP/AAPWorkshop on Periodontitisand Systemic DiseasesTonetti MS, Van Dyke TE and on behalf of working group 1 of the joint EFP/AAP workshop. Periodontitis and atherosclerotic cardiovascular disease: consensusreport of the Joint EFP/AAP Workshop on Periodontitis and Systemic Diseases.

AbstractBackground: This consensus report is concerned with the association betweenperiodontitis and atherosclerotic cardiovascular disease (ACVD). Periodontitis isa chronic multifactorial inflammatory disease caused by microorganisms andcharacterized by progressive destruction of the tooth supporting apparatus lead-ing to tooth loss; as such, it is a major public health issue.Aims: This report examined biological plausibility, epidemiology and early resultsfrom intervention trials.Plausibility: Periodontitis leads to entry of bacteria in the blood stream. The bac-teria activate the host inflammatory response by multiple mechanisms. The hostimmune response favors atheroma formation, maturation and exacerbation.Epidemiology: In longitudinal studies assessing incident cardiovascular events, sta-tistically significant excess risk for ACVD was reported in individuals with peri-odontitis. This was independent of established cardiovascular risk factors. Theamount of the adjusted excess risk varies by type of cardiovascular outcome andacross populations by age and gender. Given the high prevalence of periodontitis,even low to moderate excess risk is important from a public health perspective.Intervention: There is moderate evidence that periodontal treatment: (i) reducessystemic inflammation as evidenced by reduction in C-reactive protein (CRP) andimprovement of both clinical and surrogate measures of endothelial function; but(ii) there is no effect on lipid profiles – supporting specificity. Limited evidenceshows improvements in coagulation, biomarkers of endothelial cell activation, arte-rial blood pressure and subclinical atherosclerosis after periodontal therapy. Theavailable evidence is consistent and speaks for a contributory role of periodontitisto ACVD. There are no periodontal intervention studies on primary ACVD pre-vention and there is only one feasibility study on secondary ACVD prevention.Conclusions: It was concluded that: (i) there is consistent and strong epidemiologicevidence that periodontitis imparts increased risk for future cardiovascular disease;and (ii) while in vitro, animal and clinical studies do support the interaction andbiological mechanism, intervention trials to date are not adequate to draw furtherconclusions. Well-designed intervention trials on the impact of periodontal treat-ment on prevention of ACVD hard clinical outcomes are needed.

Maurizio S. Tonetti1, Thomas E. VanDyke2 and on behalf of working

group 1 of the joint EFP/AAPworkshop*1European Research Group on

Periodontology, Genova, Italy; 2The Forsyth

Institute, Cambridge, MA, USA

Key words: atherosclerosis; bacteremia;

cardiovascular diseases; clinical trials;

C-reactive protein; epidemiology;

inflammation; myocardial infarction;

periodontal diseases; periodontitis; stroke

Accepted for publication 14 November 2012

The proceedings of the workshop were jointly

and simultaneously published in the Journal

of Clinical Periodontology and Journal of

Periodontology.

© 2013 European Federation of Periodontology and American Academy of PeriodontologyS24

This consensus statement addressesthe association between periodontitisand ACVD. The deliberations of theWorking Group are based on a for-mal review and analysis of therecently published world literatureon the topic.

Periodontitis

Periodontitis is a chronic multifacto-rial inflammatory disease caused bymicroorganisms and characterized byprogressive destruction of the toothsupporting apparatus leading totooth loss. This is to be distinguishedfrom gingivitis. Periodontitis is amajor public health issue because(not in hierarchical order, Baehni &Tonetti 2010, Eke et al. 2012): it iscommon, it is a source of socialinequality, it reduces quality of life,it reduces chewing function andimpairs aesthetics, it causes toothloss and disability, it is responsiblefor a substantial proportion ofedentulism and masticatory dysfunc-tion, it has an impact on escalatingdental costs and it is a chronic dis-ease with possible impact on generalhealth. Periodontitis disproportion-ately affects certain groups: it ismore prevalent and severe in (i)

socially disadvantaged and specificethnic groups; and (ii) smokers, peo-ple with diabetes and the obese.

“The global burden of oral dis-eases is among the most common(non-communicable diseases). Theirimpact on individuals and communi-ties is considerable in terms of painand suffering, impairment of func-tion and reduced quality of life andcost of treatment” (FDI, World Den-tal Parliament, 2012). Article 19 ofthe recent United Nations GeneralAssembly declaration of 2011 furtherstates “… renal, oral and eye diseasespose a major health burden for manycountries and (that) these diseasesshare common risk factors and canbenefit from common responses tonon-communicable diseases.” Preser-vation of periodontal health is a keycomponent of oral and overall healthand as such is a fundamental humanright (Consensus of the EuropeanWorkshop on Periodontal Educa-tion, Baehni & Tonetti 2010).

As there are effective preventiveand treatment approaches, periodon-tal health is an attainable goal bothat the individual and at the popula-tion level. The subject of this work-shop is to assess the availableevidence whether prevention and

treatment of periodontitis has animpact on cardiovascular health.

Cardiovascular Diseases

Atherosclerotic cardiovascular dis-eases (ACVD) are a group of dis-eases that include fatal and non-fatalcoronary heart disease (angina, myo-cardial infarction), ischemic cerebro-vascular disease (stroke/TIA) andperipheral arterial disease.

The charge of the discussions ofthis group was:

1 To assess biological plausibility ofmechanisms underpinning the rela-tionship between periodontitis andcardiovascular diseases

2 To review the available epidemio-logical evidence with an emphasison longitudinal studies allowingmeasures of excess cardiovascularrisk attributable to periodontitis

3 To review the results of initialintervention trials on the benefitsof periodontal therapy on surro-gate cardiovascular outcomes

4 To critically evaluate the availableevidence spanning biological plau-sibility of mechanisms, epidemio-logical evidence and initialintervention trials

Conflict of interest and source of funding statement

Group participants declare no conflict of interest. The workshop was funded by an unrestricted educational grant from Colgate-Palmolive to the European Federation of Periodontology and the American Academy of Periodontology.

*Working group participants: James Beck, USA; Philippe Bouchard, France; Chris Cutler, USA; Francesco D’Aiuto, UK; ThomasDietrich, UK; Paul Eke, USA; Filippo Graziani, Italy; John Gunsolley, USA; David Herrera, Spain; Thomas Hart, USA; BarbaraShearer, USA; Søren Jepsen, Germany; Alpdogan Kantarci, USA; Bruno G. Loos, The Netherlands; Ann Progulske-Fox, USA;Harvey Schenkein, USA; Stefan Renvert, Sweden; Maurizio Tonetti, Italy; Thomas Van Dyke, USA; Ray Williams, USA.

© 2013 European Federation of Periodontology and American Academy of Periodontology

Periodontitis and ACVD S25

5 To identify key issues for thedesign of future trials

6 To provide reasonable action/rec-ommendations for the public, thedental and medical profession atthis stage of incomplete knowledge

Biological Plausibility

What are the biologically most plausible

mechanisms linking periodontitis to

atherothrombogenesis?

Based on the evidence outlined inthe review articles (Reyes et al. 2013,Schenkein & Loos 2013) the groupreached the following consensus ofthe most biologically plausible mech-anisms illustrated in Fig. 1. Thechronic oral infection periodontitisleads to entry of bacteria (or theirproducts) into the blood stream. Thebacteria activate the host inflamma-tory response by multiple mecha-nisms. The host immune responsefavors atheroma formation, matura-tion and exacerbation.

Is it possible to convincingly isolate

bacterial exposure from inflammatory

mediator exposure? If so, how?

It would be very challenging to dis-criminate the role of bacteria fromthe inflammatory response, but theuse of specific pharmacotherapeutic(antimicrobial, anti-inflammatory)interventions may shed light on thismatter. In animal studies, specificmediators of resolution of inflamma-tion with potent anti-inflammatoryactions block development of athero-

mas induced by periodontal bacteriaand periodontitis (Jain et al. 2003).

Is bacteremia/endotoxinemia from daily

activities and or/dental procedures more

prevalent in periodontitis patients and

associated with periodontal status?

The best available evidence suggestsa role of periodontal status in theprevalence of bacteremia after chew-ing, brushing, flossing or scaling,with a higher prevalence/incidenceand higher bio-diversity, includingperiodontal pathogens, in periodon-titis patients, versus gingivitis orhealthy patients. In addition a sys-tematic review reported an associa-tion between the prevalence ofbacteremia and plaque/gingival indi-ces (Tom�as et al. 2012).

Is there an association between

periodontal microbiota, clinical

periodontal parameters and recovery of

periodontal pathogens in atheroma

lesions?

Among the selected studies, evaluat-ing the presence of bacterial antigensand molecular signatures in the ath-erothrombotic lesions, at least tworeported a correlation between theperiodontal status (either moderateversus severe periodontitis or healthyversus periodontitis) and the pres-ence of periodontal pathogens. Inaddition, at least eight studiesdescribe a correlation between thesubgingival microbiota and thepathogens detected in the vascularlesions. There is some evidence forbacterial viability in the atheroma.

What is the role of adaptive immunity?

It is feasible that adaptive immuneresponses enhance the inflammatoryresponse in the atheroma which maylead to exacerbation. Antibodies pro-duced in response to plaque bacteriacan be pro-inflammatory, cross-react-ing with endothelial cells and withmodified LDL to enhance incorpora-tion of lipids into inflammatory cellswithin the vessel wall. Some of theseantibodies as well as inflammatorycytokines can promote Th1 responseswithin the atheroma to increase acti-vation of macrophages to enhanceinflammation in the atheroma.

Does periodontal treatment cause a short-

lived increase in systemic inflammation?

Periodontal treatment often elicits atransient increase of systemic inflam-matory/pro-thrombotic mediatorsand an overall decrease of the endo-thelial function within 24–48 hours(D’Aiuto et al. 2013). These resultsare most likely related to the bacter-emia and the trauma following thetherapeutic event.

Epidemiological Evidence

What clinical characteristics of

periodontitis have been associated with

cardiovascular disease?

The outcome in all studies in thebackground paper (Dietrich et al.2013) was incident ACVD indicatingthe occurrence of periodontitis wasprior to an incident cardiovascularevent ACVD. Therefore the discus-sions are reporting on a higher levelof evidence of association that allowsmaking statements of risk. Periodon-titis measured using clinical attach-ment loss/periodontal probing depthand/or radiographic assessment ofbone loss has been associated withincreased risk for various measuresof ACVD independent of establishedcardiovascular risk factors.

How strong are the measures of excess

risk?

Statistically significant excess risk forACVD in individuals with periodon-titis was reported to be independentof established cardiovascular riskfactors. However, the amount of theexcess risk adjusted for other ACVDrisk factors varies by type of cardio-

Figure 1. Biologically plausible mechanisms: Periodontitis and increased risk for ath-erothrombogenesis. Ath = Atheroma; B = bacteria; H = human studies; A = Animalstudies; V = in vitro studies. Dotted boxes indicate limited/no evidence.

© 2013 European Federation of Periodontology and American Academy of Periodontology

S26 Tonetti et al.

vascular outcome and across popula-tions by age and gender. Specifically,the risk is greater for cerebrovascu-lar disease than with coronary heartdisease, and greater in males and inyounger individuals. There is noexcess risk reported between mea-sures of periodontitis and incidentcoronary heart disease in subjectsolder than 65 years. This finding isconsistent with findings reported inmany studies that the strength ofestablished individual ACVD riskfactors are weaker in older adults.

There is insufficient evidence toindicate whether or not periodonti-tis is associated with the incidenceof secondary (a second ACVDevent after the original event) car-diovascular events. This finding hasimplications for future clinical trialsand, under ideal circumstances,more epidemiologic evidence wouldbe needed for the planning of suchintervention trials. Even low tomoderate excess risk reported instudies is enough to be importantfrom a public health perspectivebecause of the high prevalence ofperiodontitis.

Are there confounders that can account

for the association?

There are many potentially impor-tant confounders of the associationbetween periodontitis and ACVDrisk, including co-morbidities such asdiabetes and lifestyle factors such assmoking. However, established car-diovascular risk factors do not com-pletely explain the excesscardiovascular risk in subjects withperiodontitis. All studies included inthe review controlled for smokingstatus and excess risk was demon-strated in never-smokers in a num-ber of the studies. In studies thatcontrolled for diabetes, excess riskassociated with periodontitis wasalso demonstrated.

However, excess risk could bedue to unknown confounders.Recent findings from the ENCODEproject, a deep sequencing project toidentify all functional elements ofthe genome, indicate that there arecommon genetically determinedpathways underpinning various com-plex inflammatory diseases. There-fore, confounding by these geneticdeterminants could be due tounknown confounders.

Clinical and public health implications

Periodontal treatment requires indi-vidual professional intervention.Therefore, primary preventionbecomes more important and novelstrategies to prevent disease at thepopulation level would be highlydesirable. A diagnosis of periodontitismay contribute to cardiovascular riskstratification, if shown to improve car-diovascular risk prediction over andabove currently established predictionmodels (e.g. Framingham score).

Intervention Studies

Atherosclerotic cardiovascular diseaseis a complex multifactorial disease andindividuals may present with one or acombination of risk factors. Periodon-titis has been shown to increase therisk of future ACVD events, indepen-dent of other well-known risk factors.The group reviewed (D’Aiuto et al.2013) and graded (van Tulder et al.1997) the available evidence from peri-odontal intervention trials on ACVDoutcome.

The group concluded that withperiodontal treatment there is moder-ate evidence for reduction of systemicinflammation as evidenced by reduc-tion in CRP and improvement of bothclinical and surrogate measures ofendothelial function. Both CRP andendothelial function have been associ-ated with increased future risk of car-diovascular disease. However, there ismoderate evidence that periodontaltreatment does not have an effect onlipid profiles. There is limited evidencethat periodontal intervention reducesother ACVD biomarkers of inflam-mation, coagulation and biomarkersof endothelial cell activation. There islimited evidence that periodontaltreatment reduces arterial blood pres-sure and subclinical ACVD.

There are no periodontal interven-tion studies on primary (first ischemicevent) ACVD prevention and there isonly one feasibility study on second-ary (subsequent ischemic event)ACVD prevention. We recognize thatwell designed intervention trials onthe impact of periodontal treatmenton prevention of primary and second-ary ACVD hard clinical outcomesare needed. Two experimental designsof intervention trials can be utilized:primary ACVD prevention trials andsecondary ACVD prevention trials.

Unless surrogate ACVD outcomemeasures are used, proper controlledprimary prevention trials are unjusti-fied. Although the group recognizesthe need for additional epidemiologi-cal evidence informing clinical inter-vention trial design, secondaryACVD prevention trials should beperformed. The group recognizes anumber of challenges for designingdefinitive intervention studies. Thegroup used the PICO (population,intervention, comparison, outcome)framework to address some of thesechallenges.

How should periodontitis populations be

selected?

A number of previous trials haveincluded heterogeneity of case defini-tions. This may explain the diversityof findings from these studies. Thus,researchers should consistently adoptthe same minimal levels of severityof periodontal disease. Study popu-lations should present with substan-tial gingival inflammation (e.g.bleeding on probing or PISA scoringsystem, Nesse et al. 2008) and well-defined periodontal destruction (To-netti & Claffey 2005). Target studypopulations can be recruited frommedical offices rather than dentaloffices and we recognize that youn-ger (<65 years) study populationsare most appropriate.

What are the appropriate interventions for

periodontitis?

The group recognizes that there aremultiple effective treatment strategiesto control periodontal inflammation.Some previous studies have employedtreatment regimens that did notresolve periodontitis sufficiently. Theintervention trials should be designedbased on ACVD outcomes andinclude a pre-specified goal to elimi-nate dental biofilm and clinical gingi-val inflammation (Friedewald et al.2009) using any therapeutic strategydeemed necessary. Thus, multiplestrategies have to be employed torestore and maintain periodontalhealth in the treatment group ofRCTs. Following a ACVD event, theAHA guidelines Adams et al 2007,Jneid et al 2012 should be followed.Based on limited evidence, minimiz-ing the potential bacteremia by resto-ration of oral hygiene and staggering

© 2013 European Federation of Periodontology and American Academy of Periodontology

Periodontitis and ACVD S27

periodontal treatment in multiple ses-sions, rather than one session ofintensive treatment, is desirable.

What are the appropriate controls for

intervention trials?

A previous feasibility study (PAVE)highlighted the challenges in themanagement of subjects in the con-trol group of such studies. Controlsubjects in this particular study weretold to continue receiving their stan-dard care; however, 30% of thesecontrol patients obtained additionalperiodontal treatment, which con-founded the outcomes. Thus, futuretrials should increase sample size toaccount for such factors also takinginto consideration ethical concernsfor long-term absence of additionalperiodontal treatment.

What are the outcome measures for ACVD

that can be used in intervention trials?

The group acknowledges the chal-lenges in selection of appropriate sin-gle or composite cardiovascularoutcomes. Primary hard clinical out-comes (e.g. MI, stroke, death) arethe most relevant measures for inter-vention trials. In addition, surrogateoutcome measures (ACVD biomar-kers such as CRP, endothelial func-tion) can provide insight intomechanisms of the associationbetween the ACVD and periodontitisas reviewed in D’Aiuto et al. 2013.

What is the evidence that ACVD treatment

may impact the treatment outcomes for

periodontitis?

Emerging evidence suggests thatsome pharmacological agents maybe beneficial in reducing periodontalinflammation (e.g. aspirin, statins,fish oil, vitamin D).

Conclusions

1 There is consistent and strong epi-demiologic evidence that peri-odontitis imparts increased riskfor future ACVD

2 The impact of periodontitis onACVD is biologically plausible:translocated circulating oral mic-robiota may directly or indirectlyinduce systemic inflammation thatimpacts the pathogenesis of ath-erothrombogenesis

3 While in vitro, animal and clinicalstudies do support the interactionand biological mechanism, inter-vention trials to date are notadequate to draw further conclu-sions

Recommendations for Oral Health

Practitioners

1 Practitioners should be aware ofthe emerging and strengtheningevidence that periodontitis is a riskfactor for developing ACVD,advising patients of the risk.

2 The rationale for prevention, diag-nosis and treatment of periodonti-tis remains the preservation of thedentition and avoidance of thecrippling effects of periodontitisinduced alveolar bone loss andtooth loss.

3 Based on the weight of the evi-dence, periodontitis patients withother risk factors for ACVD, suchas hypertension, overweight/obes-ity, smoking, etc. who have notseen a physician within the lastyear, should be referred for aphysical.

4 Modifiable lifestyle associated riskfactors for periodontitis (andACVD) should be addressed inthe dental office and in the contextof comprehensive periodontaltherapy, i.e. smoking cessationprograms and advice on lifestylemodifications (diet and exercise).This may be better achieved incollaboration with appropriatespecialists and may bring healthgains beyond the oral cavity.

5 Treatment of periodontitis insubjects with a history of cardio-vascular events needs to followAHA guidelines for elective pro-cedures.

Recommendations for Research

The working group recognized thatsignificant progress has been madein understanding the relationshipbetween periodontitis and ACVD inspite of the relatively new field ofinvestigation. Significant gaps inknowledge exist and these impact onthe best way to manage patients andpopulations at risk.

More basic research is needed to:

1 Enhance understanding of bactere-mia associated with periodontaldiseases

2 Better define the role of oral mic-robiota within the atherothrom-botic lesion

3 Clarify the role inflammatorymediators produced in the peri-odontium in contributing to sys-temic host response

4 Identify genetic and epigenetic fac-tors that influence susceptibility tosystemic inflammation

5 Investigate short term vs.chronic inflammation and endo-thelial dysfunction followingperiodontal treatment in highrisk individuals

In terms of epidemiology, the fol-lowing gaps in knowledge need to beaddressed:

1 Further prospective epidemiologicstudies are needed to clarify rela-tionships between periodontitisand components of commonlyused ACVD endpoints, in particu-lar in the context of secondaryprevention studies

2 The majority of studies has usedmeasures of periodontitis ascer-tained at one point in time. There-fore, the impact of periodontalexposure over time is poorlyunderstood. Studies that look athistory of periodontitis over a per-iod of time are lacking. Thismeans that currently we do notknow whether there were anychanges in periodontal status andACVD risk factors during anylongitudinal study

3 More information is required onthe temporal relationship of theexposure to periodontitis andACVD outcome measures

4 Case-control studies (including bi-omarkers) in younger individuals(<65 years) to improve precisionof estimates are needed

With regards to intervention trials:

1 Further research is needed todefine the parameters of futurerandomized controlled trials onthe impact of treating periodonti-tis on ACVD

2 Once missing information is avail-able, well-designed interventiontrials are justified by: (i) the con-cordance of the biological plausi-

© 2013 European Federation of Periodontology and American Academy of Periodontology

S28 Tonetti et al.

bility data, the epidemiologic dataand the preliminary interventiontrials on surrogate markers; (ii)the ability to effectively treat andprevent periodontitis; and (iii) thehigh prevalence of periodontitis inthe population.

References

Adams, H. P., del Zoppo, G., Alberts, M. J.,Bhatt, D. L., Brass, L., Furlan, A., Grubb, R.L., Higashida, R. T., Jauch, E. C., Kidwell, C.,Lyden, P. D., Morgenstern, L. B., Qureshi, A.I., Rosenwasser, R. H., Scott, P. A. & Wijd-icks, E. F. M. (2007) Guidelines for the earlymanagement of adults with ischemic stroke.Stroke 38, 1655–1711.

Baehni, P. & Tonetti, M. S.; Group 1 of theEuropean Workshop on Periodontology. (2010)Conclusions and consensus statements on peri-odontal health, policy and education in Eur-ope: A call for action–consensus view 1.Consensus report of the 1st European Work-shop on Periodontal Education. European Jour-nal of Dental Education 14, Suppl 1:2–3.

D’Aiuto, F., Orlandi, M. & Gunsolley, J. C.(2013) Evidence that periodontal treatmentimproves biomarkers and ACVD outcomes.Journal of Clinical Periodontology 40 (Suppl14), 85–105.

Dietrich, T., Sharma, P., Walter, C., Weston, P.& Beck, J. (2013) The epidemiological evidencebehind the association between periodontitisand incident atherosclerotic cardiovascular dis-ease. Journal of Clinical Periodontology 40(Suppl 14), 70–84.

Eke, P. I., Dye, B. A., Wei, L., Thornton-Evans,G. O. & Genco, R. J.; CDC Periodontal DiseaseSurveillance workgroup: James Beck (Universityof North Carolina, Chapel Hill, USA), GordonDouglass (Past President, American Academy ofPeriodontology), Roy Page (University of Wash-ington). (2012) Prevalence of periodontitis inadults in the United States: 2009 and 2010.Journal of Dental Research 91,914–920.

Friedewald, V. E., Kornman, K. S., Beck, J. D.,Genco, R. J., Goldfine, A., Libby, P., Offenb-acher, S., Ridker, P. E., Van Dyke, T. E. &Roberts, W. C. (2009) The American Journalof Cardiology and Journal of PeriodontologyEditors’ Consensus: Periodontitis and athero-sclerotic cardiovascular disease. Journal of Peri-odontology 80, 1021–1032.

Jain, A., Batista, E., Serhan, C., Stahl, G. & VanDyke, T. E. (2003) A role of periodontitis inthe progression of lipid deposition in an animalmodel. Infection and Immunity 71, 6012–6018.

Jneid, H., Anderson, J. L., Wright, R. S., Adams,C. D., Bridges, C. R., Casey, D. E. Jr., Ettin-ger, S. M., Fesmire, F. M., Ganiats, T. G.,Lincoff, A. M., Peterson, E. D., Philippides, G. J.,Theroux, P., Wenger, N. K. & Zidar, J. P.(2012) ACCF/AHA focused update of theguideline for the management of patients withunstable angina/non–ST-elevation myocardialinfarction (updating the 2007 guideline andreplacing the 2011 focused update): a report ofthe American College of Cardiology Founda-tion/American Heart Association Task Forceon Practice Guidelines. Circulation 126, 875–910.

Nesse, W., Abbas, F., van der Ploeg, I., Spijker-vet, F. K., Dijkstra, P. U. & Vissink, A. (2008)Periodontal inflamed surface area: Quantifyinginflammatory burden. Journal of Clinical Peri-odontology 35, 668–673.

Reyes, L., Kozarov, E., Herrera, D., Roldan, S.& Progulske-Fox, A. (2013). Periodontal bacte-rial invasion and infection: Contribution to car-diovascular disease. Journal of ClinicalPeriodontology 40 (Suppl 14), 30–50.

Schenkein, H. A. & Loos, B. G. (2013) Inflamma-tory mechanisms linking periodontal diseasesto cardiovascular diseases. Journal of ClinicalPeriodontology 40 (Suppl 14), 51–69.

Tom�as, I., Diz, P., Tob�ıas, A., Scully, C. & Do-nos, N. (2012) Periodontal health status andbacteraemia from daily oral activities: System-atic review/meta-analysis. Journal of ClinicalPeriodontology 39, 213–228.

Tonetti, M. S. & Claffey, N.; European Work-shop in Periodontology group C. (2005)Advances in the progression of periodontitisand proposal of definitions of a periodontitiscase and disease progression for use in risk fac-tor research. Group C consensus report of the5th European Workshop in Periodontology.Journal of Clinical Periodontology 32, Suppl6:210–213.

van Tulder, M. W., Koes, B. & Bouter, L. M.(1997) Conservative treatment of acute andchronic non-specific low back pain: A system-atic review of randomized controlled trials ofthe most common interventions. Spine 22, 2128–2156.

Address:Maurizio TonettiEuropean Research Group on PeriodontologyWTC Tower Genoa, Via De Marini 1– 16149 Genova, Italy.Email: maurizio.tonetti@ergoperio.eu

© 2013 European Federation of Periodontology and American Academy of Periodontology

Periodontitis and ACVD S29