Genetic association of C-reactive protein with periodontitis: a two-sample Mendelian randomization study

Abstract

Background The association between C-reactive protein (CRP) and periodontitis has been reported in several observational studies. However, those studies are prone to residual confounding , thus limiting our understanding of the causal conclusions. Therefore, we aimed to assess the relationship between CRP and periodontitis by using the two sample Mendelian randomization (MR) method. Methods We analyzed the data from two genome-wide association studies (GWAS) of European ancestry, including CRP (204,402 individuals) and periodontitis (17353 cases and 28210 controls). We primarily used inverse variance weighted (IVW) to assess the causal effects. Pleiotropy was determined by using MR-Egger regression. Heterogeneity was examined with Cochran’s Q test. The leave-one-out analysis tested the robustness of the results. Results We obtained 54 SNPs as instrumental variables (IVs) for CRP, and these IVs had no significant horizontal pleiotropy, heterogeneity, or bias. MR analysis revealed no obvious causal effect between CRP and periodontitis (ORIVW=1.026, 95% CI 0.998–1.055, P=0.066; ORMR-Egge=1.026, 95% CI: 0.983-1.070, p = 0.242; ORweighted median=1.030, 95% CI: 0.991-1.070, p = 0.132). Conclusions The MR analysis indicated that CRP could not increase the risk of periodontitis in people of European ancestry.