Association of serum lipid levels with incident periodontitis in a multicenter study

To examine whether specific serum lipid abnormalities are associated with incident periodontitis.

We performed a retrospective, multicenter longitudinal study across 15 Korean hospitals mapped to the OMOP Common Data Model. Adults with ≥ 365 days of prior observation entered lipid-specific target or comparator cohorts on the index date: total cholesterol (TC) ≥ 200 vs. < 200 mg/dL; LDL-cholesterol (LDL) ≥ 130 vs. < 130 mg/dL; triglycerides (TG) ≥ 150 vs. < 150 mg/dL; and HDL-cholesterol (HDL) < 40 vs. ≥ 40 mg/dL. Within each hospital, 1:1 propensity-score matching balanced covariates. A 365-day lag preceded outcome surveillance to mitigate reverse causation. Incident periodontitis was identified using standardized diagnosis/procedure codes. Hospital-level Cox hazard ratios (HRs) were pooled using random-effects meta-analysis; negative-control outcomes with empirical calibration were conducted.

Periodontitis cases were 533 vs. 492 in TC cohorts (pooled HR, 1.66; 95% CI, 1.38-1.99), 1,287 vs. 1,263 in LDL (1.37; 1.20-1.57), 1,535 vs. 1,452 in TG (1.27; 1.15-1.40), and 1,224 vs. 1,056 in HDL (1.39; 1.18-1.64). Calibration analyses did not indicate substantial residual systematic bias.

In this large multicenter observational study, abnormal serum lipid profiles were associated with higher hazards of incident periodontitis. These findings support a possible link between lipid metabolism and periodontitis, but should be interpreted in light of the observational design.

Serum lipid profiles may be associated with incident periodontitis and could be considered in future studies of periodontal risk assessment and integrated medical-dental care.