Mendelian Randomization Reveals No Causal Association Between Periodontitis and Infective Endocarditis

Abstract

Objectives: Clarifying the uncertain causal relationship between periodontitis and infective endocarditis using Mendelian randomization analysis, given their historically perceived association and clinical significance.

Methods: Genetic variation data for acute periodontitis, chronic periodontitis, aggressive periodontitis, and infective endocarditis were obtained from published GWAS in individuals of European ancestry. Instrumental variables significantly associated with periodontitis were selected and univariable Mendelian randomization was conducted to infer the causal association between periodontitis and infective endocarditis. Multivariable Mendelian randomization was also performed to adjust for potential confounders including smoking, drinking, diabetes, and education.

Results: Our analysis found no evidence of a causal association between periodontitis and infective endocarditis, with odds ratios (ORs) of 0.992 (95% CI: 0.879-1.120), 0.947 (95% CI: 0.738-1.214), and 1.056 (95% CI: 0.916-1.217) for acute periodontitis, chronic periodontitis, aggressive periodontitis, respectively. The robustness of our findings was confirmed by heterogeneity tests, pleiotropy tests, leave-one-out analyses, and MR-PRESSO. In the multivariable MR analysis, adjusting for smoking, drinking, diabetes, and education, the overall patterns between genetic liability to periodontitis and infective endocarditis remained consistent (all P > .05).

Conclusion: Our findings indicate that there is no genetic causal association between periodontitis and infective endocarditis.

Keywords: Infective endocarditis; Mendelian randomization; Periodontitis.

Fig. 1

Procedure for an MR analysis of causal associations between periodontitis and infective endocarditis.

Fig. 2

The univariable MR analysis between periodontitis and IE visualized in forest plot. AP: acute periodontitis, CP: chronic periodontitis, AgP: aggressive periodontitis.

Fig. 3

The multivariable MR analysis between periodontitis and IE visualized in forest plot. AP: acute periodontitis, CP: chronic periodontitis, AgP: aggressive periodontitis.

Fig. 4

Scatter and leave one out plot of the MR analysis between periodontitis and IE. In scatter plot, the slope of each line corresponding to the causal estimates for each method. Individual SNP-effect on the outcome (point and vertical line) against its effect on the exposure (point and horizontal line) was delineated in the background. In leave one out plot, individual line represents the OR value and its 95% confidence interval after removing this SNP. (A) Scatter plot of MR analysis between AP and IE. (B) Scatter plot of MR analysis between AP and IE. (C) Scatter plot of MR analysis between AgP and IE. (D) Leave one out plot of the MR analysis between AP and IE. (E) Leave one out plot of the MR analysis between CP and IE. (F) Leave one out plot of the MR analysis between AgP and IE.