Increased serum PCSK9, a potential biomarker to screen for periodontitis, and decreased total bilirubin associated with probing depth in a Japanese community survey

Background and objectives: Previous reports suggest that several serum biomarkers play roles in the pathogenesis, inflammatory response, and oxidative stress in periodontitis caused by bacterial infections, linking chronic periodontitis to atherosclerotic vascular disease (ASVD). The aim of this preliminary study was to investigate, in a Japanese cross-sectional community survey, potential serum biomarkers of periodontitis that are associated with ASVD and chronic periodontitis.

Material and methods: The study cohort included a total of 108 male subjects who underwent annual health examinations. Serum biomarkers (high-sensitivity C-reactive protein [hs-CRP], proprotein convertase subtilisin/kexin type 9 [PCSK9], interleukin-6, tumor necrosis factor-α, soluble CD14, myeloperoxidase, matrix metalloproteinase-3, adiponectin, total bilirubin [TBIL], and serum lipids) were analyzed to determine their association (if any) with periodontal parameters. Aortic stiffness was evaluated using the brachial-ankle aortic pulse wave velocity (PWV) index and the cardio-ankle vascular index (CAVI).

Results: The concentrations of PCSK9 and hs-CRP were increased (P = .001 and .042, respectively), and the concentration of TBIL was decreased (P = .046), in subjects with periodontal disease (determined as a probing depth of ≥4 mm in at least one site) compared with periodontally healthy subjects. The ratio of low-density lipoprotein cholesterol (LDL-C) to high-density lipoprotein cholesterol and the concentrations of triglycerides, remnant-like particles-cholesterol, and oxidized LDL were elevated in subjects with periodontal disease compared with periodontally healthy subjects (P = .038, .007, .002, and .049, respectively). Multivariate regression analyses indicated that the number of sites with a pocket depth of ≥4 mm was associated with the concentration of PCSK9 and inversely associated with the concentration of TBIL independently (standardized β = .243, P = .040; standardized β = -.443, P = .0002; respectively). Analysis of receiver operating characteristic curves of PCSK9 indicated moderate accuracy for predicting the presence of disease sites (probing depth ≥ 4 mm) (area under the curve = 0.740). No significance in the values of PWV and CAVI was observed between subjects with periodontal disease and periodontally healthy subjects.

Conclusion: In Japanese male subjects, the concentrations of serum PCSK9 and TBIL were correlated with periodontal parameters. Moreover, PCSK9 could be a candidate biomarker for diagnosing chronic periodontitis, and may also have potential to evaluate the risk for periodontitis to cause ASVD. Longitudinal studies of larger populations are necessary to confirm the exact association of periodontitis with increased serum PCSK9 and decreased TBIL.