Skip to main content
Toggle Menu of ADA WebSites
ADA Websites
Partnerships and Commissions
Toggle Search Area
Toggle Menu
e-mail Print Share

A meta-analysis of observational studies reveals limited evidence of an association between periodontal and cardiovascular disease.

Lorena Baccaglini DDS, DDS, MS, PhD .


Systematic Review Conclusion

Periodontal disease may be associated with higher odds and risk of ischemic heart disease, but clinical trials are needed.

Critical Summary Assessment

This conclusion is based on a meta-analysis of 29 heterogeneous observational studies, most of which were of poor quality.

Evidence Quality Rating

Limited Evidence

Structured Abstract

Clinical Questions:

Do adults with periodontal disease or missing teeth have a greater risk of ischemic heart disease than adults without periodontal disease or missing teeth?

Review Methods:

The authors searched seven databases for observational studies published from 1989 to 2007 and containing odds ratio (OR) or relative risk (RR) estimates of cardiovascular disease in subjects with binary clinical measurements related to periodontal disease. To detect unpublished studies, they contacted authors who published two or more papers on the query topic. They selected the largest study among those using the same pool of subjects. Two examiners independently assessed study quality using published checklists (Paul Sabatiers, Sign and Cho) and resolved disagreements by discussion. The meta-analysis was conducted following the MOOSE guidelines (Stroup DF et al., 2000). Primary outcomes were coronary artery disease, myocardial infarction, angina or death from cardiopathy. Pocket depth, attachment loss and missing teeth, were used as primary outcome measures. In case of severe statistical heterogeneity, the reviewers excluded the most influential studies and used random effects models. Publication bias was tested using Begg's test at 10 percent alpha. They pooled data to calculate RR (cohort studies) and OR (case-control and cross-sectional studies) with 95 percent confidence intervals using R software (version 2.2.0). They also performed subgroup analyses by geography and periodontal or cardiovascular variables. Effects of age and study quality were evaluated by meta-regression.

Main Results:

The authors selected 32 articles from the 1,413 references. The case-control and cross-sectional studies (25/32) had significant heterogeneity, and nine of them (9/25) had poor internal validity. Three studies that yielded the most heterogeneity were excluded. Of the remaining 22 studies (19,650 subjects) the pooled OR was 2.35; 95 percent CI, 1.87 to 2.96; P < 0.0001. Of the 7 cohort studies (147,821 subjects) the pooled RR was 1.34; 95 percent CI, 1.27 to 1.42; P < 0.0001. There was no evidence of publication bias.


Subjects exposed to periodontal disease had significantly higher pooled odds and risks of developing or having cardiovascular disease than subjects without periodontal disease. However, substantial clinical and statistical heterogeneity led to the exclusion of some studies as well as difficulty comparing studies. Results should be cautiously interpreted. Furthermore, not all studies (70 percent) showed significant findings, and a dose-response effect could not be measured. There is a need for standardized clinical research protocols and improved control of confounding variables.

Source of Funding:

None disclosed.


Importance and Context:

A direct causal link between periodontitis and cardiovascular disease would impact treatment decisions for patients who are at risk for developing or who have periodontal disease. However, such a link has not been conclusively established or refuted. Meta-analyses of observational studies can help explore potential associations. However, because observational studies do not control for unmeasured confounders and adjust incompletely for measured confounders, meta-analyses based on these studies often contain substantial bias and their results should be interpreted with caution.

Strengths and Weaknesses of the Systematic Review:

This meta-analysis included a comprehensive search strategy, standardized study selection and quality assessment criteria, according to published checklists and MOOSE guidelines. The authors assessed publication bias and statistical heterogeneity. Major limitations were the lack of specificity of the clinical question and pooling of results from highly clinically heterogeneous studies. Importantly, other cohort studies (measuring hazard or rate ratios) that did not find an association were excluded (Howell et al., 2001; Hujoel et al., 2000).

Strengths and Weaknesses of the Evidence:

The 29 observational studies (167,471 participants), represent a large but biased sample of all observational studies published on this topic. A number of included studies were of poor quality. Also, some of these studies used surrogate measurements like missing teeth, which may not accurately reflect periodontitis (e.g., partially edentulous patients have fewer periodontally involved teeth and teeth could be missing for nonperiodontal reasons). Only four of seven cohort studies showed a significantly increased RR; of those studies, none measured pocket depth or clinical attachment loss. Medications, health behaviors and genetic risk factors that could lead to spuriously higher RR and OR were not adjusted for in individual studies. Thus, results should be interpreted with caution.

Implications for Dental Practice:

Although this meta-analysis of observational studies showed a modest association between periodontal disease and cardiovascular disease, results should not be interpreted as proof of a causal link due to multiple sources of bias, residual confounding and failure of the stronger cohort study designs to consistently demonstrate a significant association between pocket depth and subsequent risk of cardiovascular events. Most importantly, other cohort studies not included here do not support this association (Howell et al., 2001; Hujoel et al., 2000). Other American Dental Association’s evidence-based reviewers also noted the limited evidence of a causal association produced by similar meta-analyses (Rethman, 2009). References Stroup DF, Berlin JA, Morton SC, Olkin I, Williamson GD, Rennie D, Moher D, Becker BJ, Sipe TA, Thacker SB. “Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.: JAMA. 2000 Apr 19;283(15):2008-12. Howell TH, Ridker PM, Ajani UA, Hennekens CH, Christen WG. “Periodontal disease and risk of subsequent cardiovascular disease in U.S. male physicians. “ J Am Coll Cardiol. 2001 Feb;37(2):445-50. Hujoel PP, Drangsholt M, Spiekerman C, DeRouen TA. “Periodontal disease and coronary heart disease risk.” JAMA. 2000 Sep 20;284(11):1406-10. Chronic periodontitis associated with increased incidence of coronary heart disease Critical Summary Prepared by: Michael P Rethman DDS MS “ A Critical Summary of: Periodontal disease and coronary heart disease incidence: a systematic review and meta-analysis Humphrey LL, Fu R, Buckley DI, Freeman M, Helfand M.” J Gen Intern Med. 2008;23(12):2079-86

Critical Summary Publication Date:


These summaries are not intended to, and do not, express, imply, or summarize standards of care, but rather provide a concise reference for dentists to aid in understanding and applying evidence from the referenced systematic review in making clinically sound decisions as guided by their clinical judgment and by patient needs. American Dental Association © 2018