A cross-sectional study exploring relationships between triglyceride glucose index, atherogenic index of plasma, and
Zi Yan1,2, Hongyu Zhang3, Shumei Liu2
1Mudanjiang Collaborative Innovation Center for Development and Application of Northern Medicinal Resources, Mudanjiang, 157000, China.
Background:
Triglyceride glucose (TyG) index and atherogenic index of plasma (AIP) are indicators of insulin resistance. However, inadequate evidence indicates that the TyG index, AIP, and chronic pain are linked.
Methods:
Data from the National Health and Nutrition Examination Survey (NHANES) between 1999 and 2004 were used. Directed acyclic graphs were used to identify 11 potential confounders. The TyG index and AIP were treated as continuous variables in the multivariate logistic regression models to assess their association with chronic pain. Furthermore, the nonlinear relationships between these indices and outcomes were investigated using restricted cubic spline plots. Subsequently, subgroup analyses were conducted for the sensitive populations. Receiver operating characteristic curves were used to compare the relationships between indices and outcomes. Ultimately, two sensitivity analyses were performed.
Results:
This study identified nonlinear associations between the TyG index, AIP, and chronic pain in 16,996,513 Americans. The odds ratio and 95% confidence interval for the TyG index (per 1 standard deviation increase) was 1.17 (1.02, 1.33), and for AIP (per 1 standard deviation increase) was 1.19 (1.07, 1.34). According to the subgroup analyses, the relationships between exposure and outcome were more pronounced in the non-diabetic population. The TyG index and AIP performed similarly in assessing chronic pain in ROC curves. Additionally, the results of the two sensitivity analyses matched the conclusions of the main study.
Conclusions:
Nonlinear correlations between the TyG index, AIP, and chronic pain were identified among adults in the United States. This demonstrated that the TyG index and AIP displayed similar effectiveness in predicting the risk of chronic pain.
