Periodontal Health and Cardiovascular Disease Risk: Association or Causation?
Poor periodontal health has been positively correlated with cardiovascular disease (CVD) in numerous observational studies;1-4 however, it is not clear if poor periodontal health causes CVD. Large measures of association reported in early case-control and cross-sectional studies evaluating this association have been attributed to study design and inadequate control of confounding; these associations are attenuated but persist in meta-analyses of well-conducted cohort studies.5 Additionally, in observational studies periodontal microorganisms have been associated with atherosclerosis,6 hypertension,7 and dyslipidemia.8
Periodontal microorganisms present in dental plaque appear in the bloodstream in direct proportion to the severity and extent of periodontal infection and can impact distant sites. They have been hypothesized to cause systemic inflammation,9 impair insulin action,10 increase glycogenesis in the liver,11 and alter gut microflora resulting in systemic inflammation and metabolic changes.12 Porphyromonas gingivalis counts are higher in sub-gingival plaque in periodontal disease, where it causes inflammation and periodontal destruction.13 The organism has been shown to enter the bloodstream following tooth brushing,14 and has been found in the intima of distant blood vessels including in coronary arteries,15 where it proliferates, initiating an inflammatory cascade leading to apoptosis and consequent endothelial dysfunction,16,17 which is associated with hypertension.18 It also stimulates foam cell production in the intima leading to atherosclerosis.19 P. gingivalis has been shown to increase low-density lipoprotein (LDL) and total cholesterol levels by up-regulation of a protein which impacts circulating levels of LDL cholesterol in mice.20
There are two plausible underlying causal structures that can explain these findings: first, that periodontal disease (PD) causes CVD, and second, that PD and CVD have some unmeasured common cause (U), as depicted in Figure 1a and 1b respectively. The unmeasured common cause (U) could be an inadequately measured risk factor (such as smoking) or some combination of unmeasured genetic polymorphisms that predispose individuals to both PD and CVD. The effect of periodontal treatment (PT) in each of these scenarios is depicted in Figure 1c and 1d. If the true underlying causal relation between PD and CVD is as described in Figure 1a, then periodontal treatment will improve periodontal status and reduce CVD risk as shown in Figure 1c. If on the other hand the true underlying relation is as described in Figure 1b, then PT will improve periodontal status but will have no impact on CVD, as shown in Figure 1d. Therefore, it is possible to empirically test for the underlying causal structures linking PD and CVD by evaluating the effect of periodontal treatment on CVD.
Figure 1: Directed Acyclic Graphs (DAGs) Describing the Relation Between Periodontal Disease and CVD
In clinical trials periodontal treatment has reduced the number of pathogenic microorganisms in dental plaque,21 systemic levels of interleukin-6, C-reactive protein, blood pressure, total cholesterol, and E-selectin,22,23 and improved endothelial function,23 atherosclerotic profile,24 and glycemic control in individuals with diabetes.25 Even though these studies are consistent with the hypothesis that periodontal treatment has a beneficial effect on CVD risk factors, and by extension CVD risk, the effect of PT on CVD has not been evaluated in a randomized controlled trial.
There are several possible limitations to effectively evaluate the results of PT on CVD outcomes. 1) The effect size of periodontal treatment on CVD risk is likely small, which would require a randomized control trial (RCT) with a very large sample size to detect. 2) A single episode of periodontal treatment is unlikely to effectively control chronic periodontitis during the time it takes for CVD to develop. Long-term periodontal treatment consists of initial treatment subsequent to diagnosis, followed by regular periodontal maintenance visits. Initial treatment typically consists of non-surgical scaling and root planning, followed by 3-6 monthly periodontal maintenance. Evaluating long-term periodontal treatment in an RCT can be challenging. For example, it is possible to randomize an individual to either receive or not receive periodontal treatment at the start of a study. However, when this person returns for a periodontal maintenance visit, the dentist makes an assessment of periodontal treatment needs based on the clinical state at that time. Because the treatment that is provided following the periodontal maintenance visit is based on the dentist's judgment, PT may no longer be randomly assigned. For example, the dentist would treat if periodontal status had deteriorated, but would not treat if periodontal status was stable. PT following the maintenance visit can thus be biased due to confounding by indication.
Intention to treat analysis of PT on CVD using a conventional randomized controlled trial with treatment allocation at baseline will measure the effect of initial treatment on future CVD risk. However, this design imposes several limitations and restrictions. First, individual treatment plans vary by factors such as age, severity of periodontal disease at baseline, smoking, and medical conditions. Applying exclusion criteria based on these factors to be able to prescribe standardized treatment limits generalizability of the findings. A null finding may either mean that PT has no effect on CVD, or a group of people among whom the intervention may be effective is excluded. Second, it would be unethical to restrict periodontal treatment to the control group because periodontal treatment is effective in controlling periodontal disease. At a minimum, participants in the control group would have to be randomized to usual care. A pilot study evaluating the effect of periodontal treatment on oral health status among individuals with pre-existing CVD reported improvements in oral health parameters at six months among participants overall, but no differences between the intervention and control groups; this was ascribed to a large number of participants in the control group seeking dental care in the community.26 Third, the differences in actual treatment received between the intervention and control groups may be slight because participants in the control group would likely receive similar care from their providers. To overcome this limitation, and increase the contrast between the intervention and control groups, investigators have used intensive treatment in the intervention group.23 A drawback of this approach is that those methods are not recommended to control periodontal disease, and are generally not used in usual practice. Finally, this approach cannot evaluate the effect of long-term periodontal care on CVD because of difficulty with maintenance of long-term random assignment of PT and potential bias due to confounding by indication, as discussed earlier.27
An alternative approach is to evaluate the effect of long-term periodontal treatment (cumulative effect) in relation to systemic outcomes by using causal inference methods.28-30 Briefly, in this method inverse probability of treatment weights are calculated, which are then used to create a pseudo-population in which the confounder (or vector of multiple confounders) does not predict the exposure. In the present example, periodontal treatment in the pseudo-population would be independent of periodontal maintenance and other potential confounders, and therefore not be affected by confounding by indication. We used this approach to evaluate the effect of long-term periodontal treatment on glycemic control among individuals with diabetes in a large cohort of US veterans receiving dental care over several years.31 We identified a small beneficial effect of the intervention in the entire population, and a larger effect in the subgroup of participants with glycated hemoglobin (HbA1c) ≥9 at baseline.31 A conventional randomized controlled trial evaluating the same question did not detect these effects as it was underpowered to detect a small effect, and it excluded the subgroup with HbA1c ≥9 at baseline.32
In summary, evidence from observational data suggests that there is a small positive association between periodontal disease and CVD. Mechanistic studies and clinical trials with CVD risk factors as outcomes are consistent with the hypothesis that periodontal disease may cause CVD. This hypothesis can be tested by evaluating the effect of periodontal treatment on CVD incidence, but operationalizing this study poses unique challenges. It may be possible to get closer to the answer to this question using causal inference methods.
- Humphrey LL, Fu R, Buckley DI, Freeman M, Helfand M. Periodontal disease and coronary heart disease incidence: a systematic review and meta-analysis. J Gen Intern Med 2008;23:2079-86.
- Janket SJ, Baird AE, Chuang SK, Jones JA. Meta-analysis of periodontal disease and risk of coronary heart disease and stroke. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2003;95:559-69.
- Lafon A, Pereira B, Dufour T, et al. Periodontal disease and stroke: a meta-analysis of cohort studies. Eur J Neurol 2014;21:1155-61.
- Mustapha IZ, Debrey S, Oladubu M, Ugarte R. Markers of systemic bacterial exposure in periodontal disease and cardiovascular disease risk: a systematic review and meta-analysis. J Periodontol 2007;78:2289-302.
- Khader YS, Albashaireh ZS, Alomari MA. Periodontal diseases and the risk of coronary heart and cerebrovascular diseases: a meta-analysis. J Periodontol 2004;75:1046-53.
- Orlandi M, Suvan J, Petrie A, et al. Association between periodontal disease and its treatment, flow-mediated dilation and carotid intima-media thickness: a systematic review and meta-analysis. Atherosclerosis 2014;236:39-46.
- Desvarieux M, Demmer RT, Jacobs DR, et al. Periodontal bacteria and hypertension: the oral infections and vascular disease epidemiology study (INVEST). J Hypertens 2010;28:1413-21.
- Jaramillo A, Lafaurie GI, Millan LV, et al. Association between periodontal disease and plasma levels of cholesterol and triglycerides. Colomb Med 2013;44:80-6.
- Moutsopoulos NM, Madianos PN. Low-grade inflammation in chronic infectious diseases: paradigm of periodontal infections. Ann NY Acad Sci 2006;1088:251-64.
- Hotamisligil GS, Budavari A, Murray D, Spiegelman BM. Reduced tyrosine kinase activity of the insulin receptor in obesity-diabetes: central role of tumor necrosis factor-alpha. J Clin Invest 1994;94:1543-9.
- Ishikawa M, Yoshida K, Okamura H, et al. Oral Porphyromanas gingivalis translocates to the liver and regulates hepatic glycogen synthesis through the Akt/GSK-3β signaling pathway. Biochim Biophys Acta 2013;1832:2035-43.
- Arimatsu K, Yamada H, Miyazawa H, et al. Oral pathobiont induces systemic inflammation and metabolic changes associated with alteration of gut microbiota. Sci Rep 2014;4:4828.
- Socransky SS, Haffajee AD. Periodontal microbial ecology. Periodontol 2000 2005;38:135-87.
- Tomas I, Diz P, Tobias A, Scully C, Donos N. Periodontal health status and bacteraemia from daily oral activities: systematic review/meta-analysis. J Clin Periodontol 2012;39:213-28.
- Kozarov EV, Dorn BR, Shelburne CE, Dunn WA, Progulske-Fox A. Human atherosclerotic plaque contains viable invasive Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis. Arterioscler Thromb Vasc Biol 2005;25:e17-8.
- Roth GA, Ankersmit HJ, Brown VB, Papapanou PN, Schmidt AM, Lalla E. Porphyromonas gingivalis infection and cell death in human aortic endothelial cells. FEMS Microbiol Lett 2007;272:106-13.
- Pober JS, Min W, Bradley JR. Mechanisms of endothelial dysfunction, injury, and death. Annu Rev Pathol 2009;4:71-95.
- Rossi R, Chiurlia E, Nuzzo A, Cioni E, Origliani G, Modena MG. Flow-mediated vasodilation and the risk of developing hypertension in healthy postmenopausal women. J Am Coll Cardiol 2004;44:1636-40.
- Kebschull M, Demmer RT, Papapanou PN. "Gum bug, leave my heart alone!"—epidemiologic and mechanistic evidence linking periodontal infections and atherosclerosis. J Dent Res 2010;89:879-902.
- Miyazawa H, Tabeta K, Miyauchi S, et al. Effect of Porphyromonas gingivalis infection on post-transcriptional regulation of the low-density lipoprotein receptor in mice. Lipids Health Dis 2012;11:121.
- Knofler GU, Purschwitz RE, Eick S, Pfister W, Roedel M, Jentsch HF. Microbiologic findings 1 year after partial- and full-mouth scaling in the treatment of moderate chronic periodontitis. Quintessence Int 2011;42:e107-17.
- D'Aiuto F, Nibali L, Parkar M, Suvan J, Tonetti MS. Short-term effects of intensive periodontal therapy on serum inflammatory markers and cholesterol. J Dent Res 2005;84:269-73
- Tonetti MS, D'Aiuto F, Nibali L, et al. Treatment of periodontitis and endothelial function. N Engl J Med 2007;356:911-20.
- Teeuw WJ, Slot De, Susanto H, et al. Treatment of periodontitis improves the atherosclerotic profile: a systematic review and meta-analysis. J Clin Periodontol 2014;41:70-9.
- Wang X, Han X, Guo X, Luo X, Wang D. The effect of periodontal treatment on hemoglobin a1c levels of diabetic patients: a systematic review and meta-analysis. PLoS One 2014;9:e108412.
- Offenbacher S, Beck JD, Moss K, et al. Results from the Periodontitis and Vascular Events (PAVE) study: a pilot multicentered, randomized, controlled trial to study effects of periodontal therapy in a secondary prevention model of cardiovascular disease. J Periodontol 2009;80:190-201.
- American Academy of Periodontology. Comprehensive periodontal therapy: a statement by the American Academy of Periodontology. J Periodontol 2011;82:943-9.
- Hernan MA, Brumback B, Robins JM. Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. Epidemiology 2000;11:561-70.
- Robins JM. Association, causation, and marginal structural models. Synthese 1999;121:151-79.
- Robins JM, Hernan MA, Brumback B. Marginal structural models and causal inference in epidemiology. Epidemiology 2000;11:550-60.
- Merchant AT, Georgantopoulos P, Howe CJ, Virani SS, Morales DA, Haddock KS. Effect of long-term periodontal care on hemoglobin A1c in type 2 diabetes. J Dent Res 2016;95:408-15.
- Engebretson SP, Hyman LG, Michalowicz BS, et al. The effect of nonsurgical periodontal therapy on hemoglobin A1c levels in persons with type 2 diabetes and chronic periodontitis: a randomized clinical trial. JAMA 2013;310:2523-32.
Keywords: Apoptosis, Atherosclerosis, Blood Pressure, C-Reactive Protein, Cholesterol, LDL, Chronic Periodontitis, Coronary Vessels, Dental Care, Dental Plaque, Diabetes Mellitus, Dyslipidemias, Glycated Hemoglobin A, Hypertension, Insulin, Interleukin-6, Polymorphism, Genetic, Porphyromonas gingivalis, Risk Factors, Smoking, Primary Prevention
< Back to Listings