BACKGROUND AND OBJECTIVES: The atherogenic index of plasma (AIP), defined as log10 (triglycerides/high-density lipoprotein cholesterol), serves as a biomarker for atherosclerosis and cardiovascular disease (CVD). It is also associated with conditions such as type 2 diabetes, insulin resistance, depression, and both cardiovascular and overall mortality. Serum lipids have been proven to influence serum testosterone levels, and AIP is a significant marker of lipid levels. We hypothesize that AIP may have a specific relationship with testosterone. This article aims to evaluate the correlation between AIP and total testosterone (TT), as well as testosterone deficiency (TD), among the U.S. population. METHODS: Data were collected from the National Health and Nutrition Examination Survey (NHANES) database between 2011 and 2016. This study was categorized into four groups based on the quartiles of AIP. Weighted multivariate linear regression and logistic regression were utilized to evaluate the relationships between AIP and TT, TD. Restricted cubic spline (RCS) was used to investigate the non-linear association between AIP and TT and TD. The subgroup analysis method was employed to investigate the relationships between AIP and TT, TD across various stratifications. Ultimately, the sensitivity study involved a comparison of weighted and unweighted data analyses to ascertain the stability of the conclusions. RESULTS: A total of 2,572 participants were included in the final study. After adjusting for all confounding factors, multivariate linear regression showed that AIP was independently negatively associated with TT (β = -93.42, 95%CI: -123.66, -63.18, P <
0.001), and multivariate logistic regression showed that AIP level was associated with higher risk of TD (OR = 3.45, 95%CI: 2.09, 5.69, P <
0.001). In the quartile of AIP, TT levels decreased the most (β = -74.81, 95%CI: -105.27, -44.35, p <
0.001) and the risk of TD was highest (OR = 2.89, 95%CI: 1.70, 4.93, p <
0.001). In addition, stratified analyses showed similar results in all subgroups except those with diabetes (P for interaction >
0.05 for all comparisons). The final sensitivity analysis revealed that elevated AIP were also associated with decreased TT (β = -101.74, 95%CI: -123.18, -80.3, P <
0.001) and increased incidence of TD (OR = 3.01, 95%CI: 2.17, 4.17, P <
0.001) on unweighted data. CONCLUSION: Increased levels of AIP correlate with decreased TT levels and a higher prevalence of TD. Additional research is necessary to investigate the underlying mechanisms connecting them.