Strength of Evidence Relating Periodontal Disease and Cardiovascular Disease
1Strength of Evidence RelatingPeriodontal Disease andCardiovascular DiseaseKaumudi Joshipura, BDS, MS, ScD • Christine Seel Ritchie, MD, MSPHABSTRACT The objective of this review is to assess the strength of evidence relating periodontal disease and cardiovascular disease. Cardiovascular disease typically encompasses atherosclerosis (including coronary heart disease, peripheral arterial disease, and ischemic stroke), hemorrhagic stroke, congestive heart failure, hypertension, and rheumatic heart disease. This review focuses on atherosclerosis. Periodontal disease and cardiovascular disease may be causally linked or could be explained by common risk factors. Many potential pathways for the relationship have been postulated. This article evaluates the overall body of evidence, according to the following standard causal infer- ence criteria: strength of the association, dose-response relationship, time sequence, consistency, specificity, biologic plausibility, and independence from confounding. Each criterion is reviewed as it relates to the existing literature. The overall strength of evidence for causal criteria for the relation between periodontal disease and cardiovascular disease is as follows: specificity is not important and is not established here, the magnitude and consistency of the association is stronger for stroke, there is some initial evidence for dose response, consistency is low for coronary heart disease, time sequence has been established with more evidence for stroke, and there is definitely biologic plausibility. Independence from confounding is also stronger for ischemic stroke and peripheral arterial disease. Because the underlying pathogenesis of atherosclerosis is common across the diseases, it is likely that, should additional studies show consistent associations, periodontal disease may be an important independent causal risk factor for cardiovascular disease. C ardiovascular disease (CVD) en- compasses several diseases: athero- sclerotic CVD (including coronary This article will focus on reviewing the evidence relating periodontal disease and CVD arising from atherosclerosis heart disease [CHD], peripheral arterial (CHD, PAD, and ischemic stroke). disease [PAD], and ischemic stroke), Inflammation is now recognized as hemorrhagic stroke, congestive heart playing a key role in the pathogenesis of failure, hypertension, rheumatic heart atherosclerosis. Inflammatory cells and disease, and congenital heart defects. cytokines are not only important in theKaumudi Joshipura, BDS, MS, ScDProfessor of Epidemiology • University of Puerto Rico • Medical Sciences Campus, School of Dentistry • San Juan, Puerto RicoAdjunct Faculty • Harvard School of Dental Medicine • Harvard School of Public Health • Boston, MassachusettsChristine Seel Ritchie, MD, MSPHAssociate Professor of Medicine • Birmingham-Atlanta Veterans Administration Geriatric Research Education and Clinical Center (GRECC) •University of Alabama at Birmingham • Department of Medicine • Birmingham, Alabama INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)
2 INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA Dental Exposures Poor Orall Hygiiene Poor Ora Hyg ene Periiodonttall//Cariies Per odon a Car es Denttall Procedures Den a Procedures Tootth Loss Too h Loss Otther Orall IInffecttiions O her Ora n ec ons Bacteremia Psychosocial Diet Nutrients Factors Weight Change Noncausall Conffoundiing Noncausa Con ound ng Patthway Pa hway Common Riisk Facttors:: C o m m o n R s k Fac o r s Inflammation/Vascular Injury Genettiic Prrediisposiittiion Gene c P ed spos on Age Age Smokiing Smok ng Diiabettes D ab e es Pottenttiiall CVD Outtcomes Po en a CVD Ou comes Sociioeconomiic Sttattus Soc oeconom c S a us Sttrress,, Obesiitty,, Diiett S ess Obes y D e Attheroscllerottiic A herosc ero c Otther O her Physiicall Acttiiviitty Phys ca Ac v y Corronarry Hearrtt Diisease Co ona y Hea D sease Hyperrttensiion Hype ens on Access//Use off Denttall Carre Access Use o Den a Ca e IIschemiic Sttrroke schem c S oke Hemorrrrhagiic Sttrroke Hemo hag c S oke Healltth Awarreness//Behaviiorr Hea h Awa eness Behav o Perriipherrall Vascullarr Diisease Pe phe a Vascu a D sease Congesttiive Hearrtt Faiillurre Conges ve Hea Fa u e Rheumattiic Hearrtt Diisease Rheuma c Hea D sease CVD Morrttalliitty C V D Mo a yFigure 1 The pathways relating to periodontal disease and CVD.initiation of plaque formation in the periodontal disease (ie, tooth loss) may the association between periodontalblood vessel wall but also in the mainte- lead to dietary changes, such as decreased disease and CVD as well as some causalnance and rupture of the plaque and intake of fruits and vegetables/dietary pathways. To assess the possible exis-subsequent thrombotic complications. fiber, that could subsequently affect the tence of a causal component, the majorTriggers of inflammation include smok- risk for CVD and other diseases. Also, prospective studies are reviewed in theing, diabetes, and infectious agents.1,2 those who are genetically susceptible to context of the criteria for causality pro- Several possible pathways for the systemic inflammation may demon- posed by Hill.5 Some of these criteriarelationship between periodontal dis- strate increased oral inflammation in have been challenged or have evolvedease and CVD have been postulated the form of gingivitis or periodontal over time; however, the basic criteria,(Figure 1). Periodontal disease may disease as well as increased risk of CVD. still considered a standard approach forincrease systemic levels of inflammatory Because of this complexity, it is difficult assessing causality, are defined individ-mediators and thus potentially con- to assess whether oral disease actually ually and applied to the pertinent litera-tribute to the inflammation-associated contributes to increased risk of CVD (as ture.5,6 These criteria include strengthatherosclerotic process.3 Periodontal a causal relationship) or whether oral of association, dose-response relation-pathogens may also disseminate into disease and CVD share common risk ship, time sequence, consistency, speci-the systemic circulation and localize in factors (Figure 1). This article attempts ficity, and biologic plausibility.atheromas.4 Alternatively, individuals to review the evidence to date to under- Coherence and plausibility have beenwith periodontal disease and CVD may stand the strength of the evidence and combined into the criterion of biologicshare common behaviors or have com- to gain some insight into possible plausibility because the differencesmon host responses to inflammation causality of the relationship. between the two are very subtle.6 Also,(implying a noncausal relationship). the criterion of experiment was notFor example, those most likely to prac- CAUSAL INFERENCE assessed since there is no directtice poor dental care may be most likely CRITERIA evidence to date from clinical trials andto have other behaviors that accelerate It seems likely that there could be a it is not possible to randomly allocateCVD (eg, smoking, decreased physical combination of common risk factors people to periodontal disease. Lastly, theactivity). Alternatively, sequelae of (Figure 1) that would explain some of criterion of analogy was excluded
INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA 3 TABLE 1: Summary of Prospective Studies Relating Periodontal Disease and Coronary Heart Disease NUMBER OF YEARS OF RELATIVE RISK STUDIES PARTICIPANTS POPULATION FOLLOW-UP OUTCOME (95% CI) DeStefano, 199311 10,000 NHANES I 14 CHD 1.25* (1.06, 1.48) Mattila, 199512† 214 Finnish 7 Secondary CHD 1.21* (1.14, 1.28) Joshipura, 199613 44,119 US health professionals 6 CHD 1.04 (0.97, 1.25) Beck, 199614 1,147 US veterans 18 CHD 1.5* (1.04, 2.14) Morrison, 199915 10,368 Canadians 20 Fatal CHD 1.37 (0.80, 2.35) Hujoel, 200016 8,032 US NHANES I up to 21 CHD 1.14 (0.96, 1.36) Howell, 200117 22,037 US physicians 12 Nonfatal CHD 1.12 (0.92, 1.36) Hujoel, 200218 636 US NHANES I up to 21 Secondary CHD 0.97 (0.72–1.31) Tuominen, 200319 2,518 Finnish registry, men 12 Fatal CHD 1.0 (0.6, 1.6) 2,392 Finnish registry, women 12 Fatal CHD 1.5 (0.6, 3.8) Saremi, 200520 1,372 US diabetic Pima Indians 11 Fatal CHD 2.3 (0.9, 5.8) *Statistically significant. †Exposure used was total dental index instead of periodontal disease. NHANES = US National Health and Nutrition Examination Survey. CI = Confidence interval.because, as Rothman argues, “scientists of small biases, random chance, or con- association between periodontal diseasecan find analogies everywhere,” and “the founding. However, the absence of a and CHD. However, periodontal diseaseabsence of such analogies only reflects strong association does not rule out a was significantly associated withlack of imagination or lack of evidence.”7 causal effect. increased CHD risk among subjects Some epidemiologists have proposed Many studies have evaluated the who had very few teeth. Beck and col-alternative criteria for causality. Rothman association between periodontal disease leagues14 showed a significant increasedefines a causal mechanism as a set of and CVD. Although early work by in CHD risk among those with perio-factors that are jointly sufficient to Mattila and colleagues10 deserves credit dontal disease. Three studies assessedinduce a binary outcome event, and for stimulating interest in this area of fatal CHD. The study by Morrison15 andthat are minimally sufficient (ie, under research, and there are several subse- the one by Tuominen19 did not showthe omission of just one factor the out- quent case-control and cross-sectional significant associations, but a recentcome would change).8 This definition studies with varying degrees of study by Saremi and coworkers of type 2highlights the potential complexity of methodologic rigor, only the longitudi- diabetics showed a marginal associationcausality but provides less structure for nal studies11-19 have been included in between severe periodontal disease andevaluating the effect of one condition this review (Table 1). fatal CHD, which was significant whenon another outcome. For this article as The first prospective study was by fatal CHD was combined with mortalityin the earlier review,9 the relationship DeStefano and colleagues.11 This report from diabetic nephropathy into car-between periodontal disease and CVD was based on a 14-year follow-up study diorenal mortality.20 Two studies12,18in the context of Hill’s criteria will of National Health and Nutrition evaluated secondary outcomes of CHDbe evaluated, recognizing the inherent Examination Survey participants and among subjects who already had onelimitation in any set of criteria used to demonstrated a relative risk of 1.25 (25% heart attack (Table 1). The Mattilaassess causality. increased risk) for CHD comparing study12 showed a significant relation- those participants with periodontal dis- ship, while the Hujoel study18 did not.Strength of the Association ease to those without. The Hujoel study16 Only two studies have considered theFor this criterion, Hill argues that a used the same data set as the DeStefano relationship between PAD and perio-strong statistical association is more study,11 but controlled more rigorously dontal disease,21,22 and both of themlikely to have a causal component than for confounding factors, and found no showed significantly elevated risk ofa modest association because large relationship. Joshipura and coworkers PAD among participants with perio-associations are less likely to be a result published a study13 showing no overall dontal disease (Table 2).
4 INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA For stroke, four of the six studies Tooth Loss and partially reflects antecedent periodontalconsistently showed significantly ele- Cardiovascular Disease disease (Table 3). For tooth loss andvated relative risks (Table 2).14,23-25 The Studies that focused on the relationship CHD, there are two studies that havesignificant relative risks ranged from between tooth loss and CHD have not shown any relationship,17,19 but a1.41 to 2.28 for PAD, 1.21 to 1.5 for CHD, also been considered as part of the sup- significant relationship was seen inand 1.33 to 2.8 for stroke. porting evidence because tooth loss three cohorts.15,26 In Joshipura’s 1996 TABLE 2: Summary of Prospective Studies Relating Periodontal Disease and Other Cardiovascular Disease NUMBER OF YEARS OF RELATIVE RISK STUDIES PARTICIPANTS POPULATION FOLLOW-UP OUTCOME (95% CI) Mendez, 199821 1,110 US veterans 25-30 PAD 2.28* (1.2, 4.0) Hung, 200322 51,529 US health professionals 12 PAD 1.41* (1.12, 1.77) Beck, 199614 1,147 US veterans 18 Total stroke 2.8* (1.45, 5.48) Morrison, 199915 10,368 Canadians 20 Fatal stroke 1.63 (0.72, 3.67) Wu, 200023 9,962 US NHANES I up to 21 Ischemic stroke 2.11* (1.30, 3.42) Howell, 200117 22,037 US physicians 12 Nonfatal stroke 1.10 (0.88, 1.37) Joshipura, 200324 41,380 US health professionals 12 Ischemic stroke 1.33* (1.03, 1.70) Ajwani, 200325 364 Finnish people 10 Fatal CVD 1.97* (1.01, 3.85) *Statistically significant. NHANES = US National Health and Nutrition Examination Survey. CI = Confidence interval. TABLE 3: Summary of Prospective Studies Relating Tooth Loss and Cardiovascular Disease NUMBER OF YEARS OF RELATIVE RISK STUDIES PARTICIPANTS POPULATION FOLLOW-UP EXPOSURE OUTCOME (95% CI) Morrison, 199915 4,285 Canadians 20 0 teeth CHD 1.90* (1.17, 3.10) Howell, 200117 22,037 US physicians 12 Tooth loss Nonfatal CHD 1.21 (0.80, 1.83) Tuominen, 200319 2,518 Finnish registry, men 12 0–10 teeth Fatal CHD 0.9 (0.5, 1.6) 2,392 Finnish registry, women 12 0–10 teeth Fatal CHD 0.3 (0.1, 1.0) Hung, 200426 41,407 US health professionals 12 0–10 teeth CHD 1.36* (1.11, 1.67) 58,974 US nurses 6 0–10 teeth CHD 1.64* (1.31, 2.05) Hung, 200322 45,136 US health professionals 12 Recent tooth loss PAD 1.39* (1.07, 1.82) 45,136 US health professionals 12 0 teeth PAD 1.05 (0.68, 1.63) Morrison, 199915 10,120 Canadians 20 0 teeth Fatal stroke 1.63 (0.77, 3.42) Wu, 200023 9,962 US NHANES 14 0 teeth Ischemic stroke 1.41 (0.96, 2.06) Howell, 200117 22,037 US physicians 12 Tooth loss Nonfatal stroke 1.20 (0.76, 1.89) Joshipura, 200324 44,116 US health professionals 12 0–24 teeth Ischemic stroke 1.57* (1.24, 1.98) Ajwani, 200325 364 Finnish people 10 0 teeth Fatal CVD 1.40 (0.76, 2.59) *Statistically significant. NHANES = US National Health and Nutrition Examination Survey. CI = Confidence interval.
INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA 5report,13 a significant relationship was TABLE 4:observed for the combination of tooth Results from Meta-analyses of Studies Relatingloss and periodontal disease; the relative Periodiontal Disease and CVDrisk for tooth loss was elevated but notsignificant. The subsequent report byHung with a longer follow-up found STUDIES INCLUDED IN RELATIVE RISKsignificant associations between tooth STUDY THE META-ANALYSES OUTCOME (95% CI)loss and CHD in the same cohort ofmale professionals as well as in an addi- Danesh, 199927 5 prospective CHD 1.24 (1.10–1.38)tional cohort of women.26 The relation- Muller, 200228 4 prospective CHD 1.12 (0.95–1.33)ship between PAD and recent tooth loss 3 prospective Stroke 1.73 (0.89–3.34)showed a stronger association than toothloss that occurred in the distant past. Janket, 200329 8 prospective CHD/Stroke 1.19* (1.08–1.32)That is, tooth loss in the past six years 4 prospective > CHD/Stroke (— 65 y) 1.44* (1.20–1.73)was associated with elevated risk for 2 prospective Stroke 2.85* (1.78–4.56)PAD, but the baseline number of teeth Khader, 200430 6 prospective + 2 CHD 1.15* (1.06–1.25)was not significantly associated with 4 prospective + 2 Stroke 1.13* (1.01–1.27)PAD (Table 3).22 For stroke, only onestudy showed a significant association *Statistically significant. CI = Confidence interval.for tooth loss.24 When comparing periodontal dis- studies have looked at the dose- longitudinal studies, these studies pro-ease and tooth loss, it seems that overall response relationship for stroke. In one vide much better support for causalperiodontal disease and tooth loss cross-sectional study, there was a clear inference concerning the relationshipdemonstrate similar relationships with dose response,32 but Beck’s study found between periodontal disease and CVDCHD and with stroke.22 Among health no dose-response relationship for than case-control and cross-sectionalprofessionals, recent tooth loss follows stroke.14 There is also some indirect studies. However, given the chronicitythe same pattern as periodontal disease evidence for dose response. In a cross- of both periodontal disease and CVD, it(significant for PAD22 and stroke,24 but sectional study, overall periodontal is difficult to know for sure, even in lon-not for CHD26), which may be expected bacterial burden (defined by the score of gitudinal studies, whether the perio-because if people lose teeth in their 40s Actinobacillus actinomycetemcomitans, dontal disease truly preceded the earlyand 50s, it is likely to be a result of Porphyromonas gingivalis, Tannerella stages of CVD. It seems unlikely thatperiodontal disease. forsythensis, and Treponema denticola) CVD could cause periodontal disease. In summary, the association between was significantly related to carotid Hence, there does seem to be evidence,periodontal disease and CVD appears intimal thickness.33 Increases in carotid strongest for stroke but also for CHDstronger for both PAD and stroke than intimal medial thickness (IMT), as and PAD, suggesting that the time-for CHD, as is also suggested from the measured by noninvasive ultrasonogra- sequence criterion has been established.meta-analyses (Table 4).27-30 According phy, have been associated with increasedto the “strength of association” criteria, risk of myocardial infarction and stroke, Consistencythe overall body of evidence relating particularly in adults 65 years of age If several studies show similar results,periodontal disease to CHD and PAD is or older.34 it can be said the relationship is con-weak, but stronger for stroke. sistent. Consistently finding an asso- Time SequenceDose-Response Relationship For the time sequence criterion to be ciation with different study designs andTo fulfill this criterion, the outcome met, the potential causal factor must populations reduces the likelihood thatincreases with increasing dose of expo- precede the outcome. This is best ascer- an association would be a result of asure. A dose-response relationship is tained in longitudinal studies, and ideally “constant” error in the design.not always found in causal relation- in randomized controlled trials, when it For CHD, many studies foundships, in which case a more complex is practicable and ethical to randomly insignificant results, and, overall, theexplanation of the relationship may be allocate the postulated causal factor. results were not consistent (Table 1).required.6 There are several studies in which Therefore, more CHD studies are needed Very few studies have evaluated dose the exposure clearly preceded the out- to corroborate the relationship. The rela-response. Beck and colleagues14 and come. Of the longitudinal studies to tionship is most consistent for stroke, inGeerts and colleagues (case-control date, three of the ten CHD stud- which four14,23-25 of the six studiesstudy)31 assessed dose response relating ies,11,12,14 both PAD studies, and five of found an elevated relationship. Bothincreasing levels of periodontal disease the six stroke studies showed a relation- studies on PAD show consistent results,with CHD risk and both found a signif- ship. Because periodontal disease pre- but this needs to be replicated in moreicant dose-response relationship. Two cedes the outcome (CVD) in the longitudinal studies.
6 INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA The possible explanations for theinconsistency for CHD could include Inc rea Mechance variation (some studies observed s Med a d Ischemic Stroke Tum ed C- iiattorr:: Tum C oor did not observe an association by orr N -Reac eg,, o N R eac eg ecro ttiive ecrchance). Alternatively, differences in osiis ve Prro s s F P ot actto teiin,, Fac epopulation characteristics, limitations or α n αin the studies of exposure measures, P e r i o d o n t al D i s e a s eoutcome measures, or control of con-founders may explain the inconsis-tencies. Site differences (eg, differentproximities between the heart and thebrain as it relates to the mouth) andsmall differences in arterial flow betweencerebral, coronary, and peripheralvasculature may explain some of this Confounder: (eg, Geneticinconsistency. P r e d i s p os i t i o n , A g e , S m o ki n g)Limitation inExposure Measures Figure 2 Differentiating confounders versus mediators.Periodontal exposure measures varyacross studies. Pocket depth, attachment using self reports are likely to be attenu- Population Differencesloss, and bone loss are the standard ated compared with clinical measures. There could be genuine differencespopulation-based measures for perio- Self reports were used in articles by between the populations studied, whichdontal disease. Although these are stan- Joshipura and colleagues.13,22,24 The self could lead to differences in associationsdard measures, there is still no universal reports showed good validity against if the associations only exist amongdefinition or cut-off for periodontal radiographic bone loss in populations subgroups such as younger people,disease. Therefore, the threshold is not of health professionals,35,36 including smokers, etc. The population differ-predefined and the measures vary. In dentists, who are better able to report ences in race, socioeconomic status, oraddition, the possibility exists that the periodontal status. Self reports were smoking status may explain some of theteeth with more severe periodontal also found to perform just as well as the inconsistencies. Consistency is lackingdisease were extracted; therefore, there clinical periodontal measures in assess- for CHD, but is reasonable for stroke.are limitations even in the “standard” ing a linear relationship with age.35measures. Independence Some studies use composite meas- Limitations in CVD Outcome From Confoundingures of a total dental index; however, Measures A confounder is an extraneous factor,because caries, tooth loss, and perio- CVD outcomes are also not consistent which leads to an apparent associationdontal disease are together in one index, across studies. Angina, which is a softer between the exposure and outcome thatit is difficult to distinguish which expo- measure than myocardial infarction, is different from the true association.sure is actually related to CVD out- was included in the CHD outcome in This criterion was included in our 2000comes. Tooth loss as an exposure could some studies. Some stroke studies article,9 but was not explicitly men-also be partly considered a surrogate focused on ischemic stroke, some tioned in the Hill criteria. One reasonmarker for periodontal disease because included all strokes (both hemorrhagic it was omitted may have been that it wasperiodontal disease, caries, and ortho- and ischemic), and some included tran- subsumed under the other criteria.dontic concerns can all contribute to sient ischemic attack (a transient occlu- Given the multitude of confounderspotential extractions. sion of a cerebral vessel). In addition, and complexity of adjusting for some of Self-reported measures of perio- outcomes in the CVD studies varied these, we thought it was important todontal disease have been criticized for from fatal to nonfatal to total CVD. The emphasize this criterion separately. Theproviding limited information as well as degree of verification of the CVD out- association between periodontal diseasethe inability of participants to recognize come also varies across studies. These and CVD should be independent ofsubtle changes in periodontal status. limitations must be kept in mind when confounding at least from the majorHowever, if an association is observed considering inconsistency. known risk factors for the disease. Unlessusing self reports, it is unlikely that the it can be shown that the association isbias from the measure is in the direc- Limitations in Controlling independent of common risk factors,tion of observing a stronger association. for Confounding causal interpretation is meaningless.Rather, random misclassification gener- This is addressed in the section on con- Confounders are risk factors that areally biases the estimates towards the null; founding and could possibly explain the common to the exposure and outcome,therefore, the associations observed inconsistencies in the results. and can lead to a deceptive association
INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA 7 TABLE 5: for health awareness and all but one15 Summary of Evidence Supporting Causal Criteria controlled for socioeconomic status. Relating Periodontal Disease and Cardiovascular Disease* Because the studies among health pro- fessionals were the only ones that had the opportunity to control for CARDIOVASCULAR DISEASE health awareness, positive associations HILL’S CRITERIA CHD PAD ISCHEMIC STROKE in this population provide stronger evi- dence for independence from con- Specificity — — — founding. Among health professionals, Strength of association + + ++ positive associations were found between Dose-response relationship + — + periodontal disease and ischemic stroke and PAD. Time sequence + + ++ Biologic plausibility ++ ++ ++ Specificity Consistency — + ++ Specificity is one of the criteria postu- lated by Hill, but many do not regard it Independence from confounding + ++ ++ as important. It can be established when *Scale: — — — to + + +. a single putative cause produces a spe- The consistency of the CHD research was discussed and determined to be suggestive; however, there is more inconsistency in these studies than the other conditions being considered. cific effect. This is not true in CVD, as in Many more studies have been conducted on CHD and diabetes than on PAD or ischemic stroke. many other diseases, because we know there are multiple factors that increasebetween two factors (Figure 2). If a Danesh criticized several studies for not risk. Specificity provides additionalstudy finds an association between adequately controlling for socio- support for causality, but absence ofperiodontal disease and CVD, it may economic status.38 In addition to fac- specificity (multiple causes) as in CVDmean that periodontal disease causes tors that can be measured and con- does not negate a causal relationship.CVD, but it may just mean that com- trolled, there are factors that are hard tomon risk factors could cause both. For measure and hard to control for, such as Biologic Plausibilityexample, age is a confounder, smoking health awareness or health behavior. Ideally, the observed association shouldis a confounder, and they both increase Studies focusing on relatively homoge- be biologically explainable and shouldthe risk for both the periodontal expo- neous populations, such as health pro- not completely contradict the overallsure and the CVD outcome. Because of fessional groups, are able to partly scientific knowledge. Once a statisticalconfounders, a relationship could be control for such factors. Hujoel and col- relationship is found, it needs to beapparent between periodontal disease leagues tabulated the degree of control determined if it is biologically plausible.and ischemic stroke even if there was no of confounding by smoking dose and Generally, if the epidemiologic associa-causal relation. In the design of a study, health awareness in various prospective tions are established, causality is moreone good way to control for confound- studies.37 In the studies by Joshipura, likely if a supported biologic explana-ing is randomization. Clinical trials are Hung, and Howell, health awareness tion exists for it. There are many poten-advantageous because, if randomized and behavior were at least partially con- tially biologically plausible explanationsand sufficiently large, they are likely to trolled for because the sample included for the relationship between perio-be free of confounding. Often random- groups of health professionals. These dontal disease and CVD (Figure 1).ization is not feasible or is too costly. In professionals knew more about health Chronic infection may initiate athero-these instances, the optimal observa- and therefore might have been likely to sclerosis or interact with other risktional study will control for all of the do more to prevent CVD as well as to factors to amplify the vessel inflamma-important confounding variables. Over- prevent oral disease. Among the ten tory response.39 This response may becontrolling occurs if the study analysis prospective studies relating periodontal manifested by alteration of endothelialcontrols for mediators. Mediators are disease and CHD shown in Table 1, two function or acceleration of plaque for-part of the biologic pathway, but con- control for health awareness13,17 and mation. Acute infection may destabilizefounders are just common risk factors. seven control for socioeconomic status plaques or exert inflammatory andA mediator is a step in the causal path- in some manner.11-13,16-19 The study by thrombotic effects on atheroscleroticway and occurs in time between the Saremi and colleagues20 did not directly plaques.40 Infection may also contributeexposure and outcome (Figure 2). control for socioeconomic status but to elevation of acute phase proteins,Differentiating mediators and con- free dental and medical care was avail- which may in turn modulate atherogen-founders can often be very difficult and able to all participants. Both PAD stud- esis.41 Many studies demonstrate anneeds an understanding of the biology. ies21,22 controlled for socioeconomic association between periodontal disease Hujoel and others have emphasized status, one of which also controlled for and acute phase proteins such as C-the importance and difficulty of ade- health awareness. Among the six stroke reactive protein and fibrinogen.42,43quately controlling for smoking.37 studies in Table 2, two17,24 controlled Studies have also demonstrated the
8 INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURApresence of oral pathogens in arterial However, it is important to note that 6. Hofler M. The bradford hill considerationsplaque. In the study by Haraszthy and clinical trials would not be able to on causality: a counterfactual perspective.colleagues, 44 of surgical specimens answer all the questions. Because of Emerg Themes Epidemiol. 2005;2:11.obtained during carotid endarterectomy, practical considerations, there are diffi- 7. Rothman KJ, Greenland S. Modern epi-44% of the 50 atheromas were positive culties, such as how many periodontal demiology. Philadelphia, PA: Lippincott-for at least one of the target periodontal treatments to allocate to make the two Raven;1998.pathogens. In the study by Beck and groups sufficiently different and how 8. Rothman KJ. Causes. Am J Epidemiol.coworkers,45 of IMT, participants with long a follow-up period is feasible to 1976;104:587-592.antibodies to specific periodontal patho- enable accumulation of sufficient num- 9. Joshipura K, Ritchie C, Douglass C. Strengthgens had a great likelihood of having ber of cardiovascular cases while limit- of evidence linking oral conditions andincreased IMT. Of particular relevance ing attrition. Clinical trials would also systemic disease. Compend Contin Educto the theory that periodontal-disease– not be able to provide direct informa- Dent. 2000;30(suppl):12-23.induced inflammation alters endothelial tion on pathophysiology. More impor- 10. Mattila KJ, Nieminen MS, Valtonen VV,function, is the recent report demon- tantly, trials can only compare people et al. Association between dental healthstrating improvement in endothelial with and without periodontal treat- and acute myocardial infarction [see com-function in patients after treatment of ment; whereas, only observational stud- ments]. BMJ. 1989;298:779-781.periodontal disease.46 ies can suggest means for prevention by 11. DeStefano F, Anda RF, Kahn HS, et al. comparing CVD risk between people Dental disease and risk of coronary heartOVERALL STRENGTH with and without periodontal disease. disease and mortality. BMJ. 1993;306:OF EVIDENCE Hence, a combination of observational 688-691. and intervention studies is needed. 12. Mattila KJ, Valtonen VV, Nieminen M, et al.Table 5 summarizes the overall strength Dental infection and the risk of new coro-of evidence according to the causal cri- nary events: prospective study of patientsteria for CVD. In summary, the overall CONCLUSION with documented coronary artery disease.strength of evidence for causal criteria At the present time, there is insufficient, Clin Infect Dis. 1995;20:588-592.for the relation between periodontal but suggestive, evidence for a possible 13. Joshipura KJ, Rimm EB, Douglass CW,disease and CVD is as follows: causal relation between periodontal dis- et al. Poor oral health and coronary heart • Specificity is not important and is ease and CVD, with slightly stronger disease. J Dent Res. 1996;75:1631-1636. not established here. evidence for stroke. If future studies 14. Beck J, Garcia R, Heiss G, et al. Perio- • The magnitude and consistency of show consistent associations, perio- dontal disease and cardiovascular disease. dontal disease may be elucidated as an the association is stronger for stroke. J Periodontol. 1996;67:1123-1137. independent and potentially modifiable • There is some initial evidence for 15. Morrison HI, Ellison LF, Taylor GW. causal risk factor for CVD. dose response. Periodontal disease and risk of fatal coro- • Consistency is low for CHD. nary heart and cerebrovascular diseases. ACKNOWLEDGMENT J Cardiovasc Risk. 1999;6:7-11. • Time sequence has been established The authors would like to thank Dr. with more evidence for stroke. 16. Hujoel PP, Drangsholt M, Spiekerman C, Chester Douglass, Dr. Walter Willett, et al. Periodontal disease and coronary heart • There is definitely biologic plausibility. NIDCR R01DE12102, K24DE016884, Independence from confounding is disease risk. JAMA. 2000;284:1406-1410. and the Office of Dietary Supplements. 17. Howell TH, Ridker PM, Ajani UA, et al.also stronger for ischemic stroke and Periodontal disease and risk of subse-PAD. The biologic links for the associa- REFERENCES quent cardiovascular disease in US maletion between periodontal disease and 1. Tiong AY, Brieger D. Inflammation and physicians. J Am Coll Cardiol. 2001;37:CVD are discussed in more detail by coronary artery disease. Am Heart J. 2005; 445-450.Seymour and colleagues in a separate 150:11-18. 18. Hujoel PP, Drangsholt M, Spiekerman C,article related to the Global Oral Health– 2. Fuster V, Moreno PR, Fayad ZA, et al. et al. Pre-existing cardiovascular diseaseSystemic Health Forum.47 Atherothrombosis and high-risk plaque: and periodontitis: a follow-up study. J Dent Part i: evolving concepts. J Am Coll Cardiol. Res. 2002;81:186-191. 2005;46:937-954.FUTURE DIRECTIONS 19. Tuominen R, Reunanen A, Paunio M, et al. 3. Meurman JH, Sanz M, Janket SJ. OralAdditional well-conducted prospective Oral health indicators poorly predict coro- health, atherosclerosis, and cardiovascularstudies are needed to enable the assess- disease. Crit Rev Oral Biol Med. 2004;15: nary heart disease deaths. J Dent Res.ment of additional populations (includ- 403-413. 2003;82:713-718.ing developing countries), for evaluating 4. Chun YH, Chun KR, Olguin D, et al. 20. Saremi A, Nelson RG, Tulloch-Reid M, et al.the role of genetic factors, and for evalu- Biological foundation for periodontitis as a Periodontal disease and mortality in type 2ating mediators. The evidence linking potential risk factor for atherosclerosis. diabetes. Diabetes Care. 2005;28:27-32.periodontal disease and CHD, especially J Periodontal Res. 2005;40:87-95. 21. Mendez MV, Scott T, LaMorte W, et al. Anfor the independence from confounding 5. Hill AB. The environment and disease: association between periodontal diseasecriterion, would be greatly enhanced Association or causation? Proceedings Royal and peripheral vascular disease. Am Jwith randomized controlled trials. Society of Medicine. 1965;58:295-300. Surg. 1998;176:153-157.
INSIDE DENTISTRY VOL. 2 (SPECIAL ISSUE 1)—INTERNATIONAL CONSENSUS STATEMENT—KAUMUDI JOSHIPURA 922. Hung HC, Willett WC, Merchant A, et al. 35. Joshipura KJ, Douglass CW, Garcia RI, et Oral health and peripheral artery disease. al. Validity of a self-reported periodontal Circulation. 2003;107:1152-1157. disease measure. J Public Health Dent.23. Wu T, Trevisan M, Genco RJ, et al. 1996;56:205-212. Periodontal disease and risk of cere- 36. Pitiphat W, Garcia RI, Douglass CW, et al. brovascular disease: the first national Validation of self-reported oral health health and nutrition examination survey measures. J Public Health Dent. 2002; and its follow-up study. Arch Intern Med. 62:122-128. 2000;160:2749-2755. 37. Hujoel PP. Does chronic periodontitis24. Joshipura KJ, Hung HC, Rimm EB, et al. cause coronary heart disease? A review of the literature. J Am Dent Assoc. 2002; Periodontal disease, tooth loss, and inci- 133(suppl):31S-36S. dence of ischemic stroke. Stroke. 2003; 38. Danesh J, Whincup P, Walker M, et al. Low 34:47-52. grade inflammation and coronary heart25. Ajwani S, Mattila KJ, Narhi TO, et al. disease: prospective study and updated Oral health status, c-reactive protein and meta-analyses. BMJ. 2000;321:199-204. mortality—a 10 year follow-up study. 39. Kiechl S, Egger G, Mayr M, et al. Chronic Gerodontology. 2003;20:32-40. infections and the risk of carotid athero-26. Hung H, Joshipura K, Colditz G, et al. The sclerosis: prospective results from a association between tooth loss and coro- large population study. Circulation. 2001; nary heart disease in men and women. 103:1064-1070. Journal of Public Health Dentistry. 2004; 40. Hansson GK. Inflammation, atherosclero- 64:209-215. sis, and coronary artery disease. N Engl27. Danesh J. Coronary heart disease, heli- J Med. 2005;352:1685-1695. cobacter pylori, dental disease, chlamydia 41. Danesh J, Whincup P, Walker M, et al. Low pneumoniae, and cytomegalovirus: meta- grade inflammation and coronary heart analyses of prospective studies. Am disease: prospective study and updated Heart J. 1999;138:S434-437. meta-analyses. BMJ. 2000;321:199-204.28. Muller HP. Does chronic periodontitis play 42. Noack B, Genco RJ, Trevisan M, et al. a role in the pathogenesis of cardio- Periodontal infections contribute to elevated vascular and cerebrovascular diseases? systemic c-reactive protein level. J Perio- Gesundheitswesen. 2002;64:89-98. dontol. 2001;72:1221-1227.29. Janket SJ, Baird AE, Chuang SK, et al. 43. Wu T, Trevisan M, Genco RJ, et al. Meta-analysis of periodontal disease and Examination of the relation between risk of coronary heart disease and periodontal health status and cardiovas- stroke. Oral Surg Oral Med Oral Pathol cular risk factors: serum total and high Oral Radiol Endod. 2003;95:559-569. density lipoprotein cholesterol, c-reactive30. Khader YS, Albashaireh ZS, Alomari MA. protein, and plasma fibrinogen. Am J Periodontal diseases and the risk of coro- Epidemiol. 2000;151:273-282. 44. Haraszthy VI, Zambon JJ, Trevisan M, et nary heart and cerebrovascular diseases: al. Identification of periodontal pathogens a meta-analysis. J Periodontol. 2004; in atheromatous plaques. J Periodontol. 75:1046-1053. 2000;71:1554-1560.31. Geerts SO, Legrand V, Charpentier J, et al. 45. Beck JD, Elter JR, Heiss G, et al. Further evidence of the association Relationship of periodontal disease to between periodontal conditions and coro- carotid artery intima-media wall thick- nary artery disease. J Periodontol. 2004; ness: the atherosclerosis risk in commu- 75:1274-1280. nities (ARIC) study. Arterioscler Thromb32. Elter JR, Beck JD, Offenbacher S, et al. Vasc Biol. 2001;21:1816-1822. Relationship of periodontal disease and 46. Seinost G, Wimmer G, Skerget M, et al. edentulism to stroke/TIA. J Dent Res. Periodontal treatment improves endothe- 2003;82(12):998-1004. lial dysfunction in patients with severe33. Desvarieux M, Demmer RT, Rundek T, et al. periodontitis. Am Heart J. 2005;149: Periodontal microbiota and carotid intima- 1050-1054. media thickness: the oral infections and 47. Seymour GJ, Ford PJ, Gemmell E, vascular disease epidemiology study Yamazaki K. Oral and systemic health: (invest). Circulation. 2005;111:576-582. consensus statement from an international34. O’Leary DH, Polak JF. Intima-media thick- panel. Article 5: Infection or inflammation: ness: a tool for atherosclerosis imaging the link between periodontal disease and and event prediction. Am J Cardiol. 2002; systemic disease. Inside Dentistry [CD-ROM]. 90:18L-21L. 2006;2(special issue 1).