Predictive power of lipid-related indicators for testosterone deficiency: a comparative analysis, NHANES 2011-2016

Studies have shown that lipid-related indicators are associated with testosterone deficiency. However, it is difficult to determine which indicator is the most accurate predictor of testosterone deficiency. We aimed to identify the lipid-related indicators most predictive of testosterone deficiency in adults in the United States.

This observational research was conducted on a population aged ≥ 20 years. By plotting the receiver operating characteristic curve (ROC) and obtaining the corresponding area under the curve (AUC) value, we assessed the predictive capacity of TyG, WTI, LAP, and VAI for testosterone deficiency. We compared the area under the curve (AUC) values of these measures to determine if there were any statistically significant differences. The relationship between lipid-related indices and testosterone hormones was investigated using regression modeling, eXtreme Gradient Boosting (XGBoost) modeling, and sensitivity analysis.

A total of 3,272 eligible participants were included in the study. Testosterone deficiency was found to exist in 20.63% of the participants. Subjects with higher lipid-related markers were more likely to have lower testosterone levels. LAP was the best predictor of testosterone deficiency in ROC analysis over other indicators (AUC = 0.7176, (95% CI: 0.6964-0.7389)).

LAP is the most straightforward and convenient indicator for identifying testosterone deficiency in clinical practice.