Abstract
INTRODUCTION: Observational studies have indicated that individuals with hypertension (HTN) and diabetes mellitus (DM) tend to exhibit elevated plasma atherogenic index of plasma (AIP), defined as log (triglyceride [TG]/high-density lipoprotein cholesterol (HDL-C)). However, the precise relationship between these factors remains unclear. This study aimed to examine the correlations among HTN, DM, and AIP. METHODS: Data from the National Health and Nutrition Examination Survey (NHANES; 2011-2016), a nationally representative sample, were analyzed to assess the relationship between AIP and the coexistence of HTN and DM in United States (US) adults. AIP served as the exposure variable, adjusted for 28 covariates. Baseline characteristics, correlation analysis, stratified analysis, and non-linear modeling were employed to elucidate these associations. The Extreme Gradient Boosting (XGBoost) machine learning algorithm was utilized to evaluate the predictive value of various variables for the presence of HTN and DM. Receiver operating characteristic (ROC) curve analysis was performed to assess AIP's diagnostic accuracy for detecting HTN and DM. RESULTS: Baseline characteristics revealed that individuals with HTN and DM had higher mean AIP values (0.39). Participants with alcohol use, obesity, or metabolic syndrome were more likely to present with both conditions. A significant positive correlation between AIP and the coexistence of HTN and DM was found (model 1: odds ratio [OR] = 5.93, 95% confidence interval [CI] = 3.84-9.16, P < 0.001; model 2: OR = 6.78, 95% CI = 4.14-11.1, P < 0.001; model 3: OR = 3.95, 95% CI = 1.66-9.39, P = 0.005), as confirmed by stratified analysis and smoothing curve analysis. The XGBoost algorithm identified AIP as an important predictor of HTN and DM. ROC curve analysis demonstrated AIP's relatively high accuracy in predicting these conditions. Smoothing curve analysis further supported the positive associations among AIP, HTN, and DM. CONCLUSION: This cross-sectional study highlights AIP was significantly associated with HTN combined with DM, underscoring its potential as a diagnostic tool. These findings provide valuable insights for future preventive and therapeutic approaches.