Comparison of the triglyceride glucose index and modified triglyceride glucose indices in assessing periodontitis in Korean adults

BackgroundPeriodontal diseases are closely connected with insulin resistance (IR) and metabolic syndrome. The Triglyceride Glucose Index (TyG) assesses IR, and recently, a few indices combining TyG and body composition have emerged with higher IR predictive performance than TyG alone. We aimed to examine which TyG-related parameters are most helpful in predicting the risk of periodontal disease. MethodsFrom 2013 to 2015, data were collected through the Korean National Health and Nutrition Examination Survey. Periodontitis was defined using the Community Periodontal Index. TyG-body mass index (BMI), TyG-waist circumference (WC), and TyG-waist-to-height ratio (WHtR) were calculated by multiplying TyG index score by BMI, WC, and WHtR, respectively. Multiple logistic regression analysis was used to calculate odds ratio (OR) and 95% confidence intervals (CI). Receiver operating characteristic curves were used to estimate areas under the curve (AUC). ResultsCompared to each reference quartile (Q)1, Q4 of the TyG index, TyG-BMI, TyG-WC, and TyG-WHtR were significantly associated with a higher risk of periodontitis after adjusting for confounders (OR, 95% CI; 1.23, 1.01-1.49 for TyG; 1.63, 1.22-2.17 for TyG-BMI; 1.37, 1.04-1.81 for TyG-WC; and 1.53, 1.16-2.02 for TyG-WHtR). The AUC and 95% CIs of TyG, TyG-BMI, TyG-WC, and TyG-WHtR in predicting periodontitis were 0.609 (0.600-0.617), 0.605 (0.596-0.617), 0.629 (0.621-0.637), and 0.636 (0.628-0.644), respectively (all p < .001). ConclusionsTyG, TyG-BMI, TyG-WC, and TyG-WHtR appear to be significantly associated with the prevalence of periodontitis. Among the TyG and modified TyG indices, TyG-WHtR exhibited the highest predictive performance for periodontitis.

Automated summary

This study found that TyG, TyG-BMI, TyG-WC, and TyG-WHtR are significantly associated with the risk of periodontitis, and TyG-WHtR has the highest predictive performance.