The Triglycerides and Glucose Index is not superior to HOMA-IR in predicting testosterone deficiency among adult males.

BACKGROUND

Testosterone decline and deficiency in males have aroused increased attention in male health management, which might have a close relationship with insulin resistance (IR).

OBJECTIVES

We utilized a novel and practical IR indicator, the triglyceride-glucose (TyG) index, to investigate the association between the TyG index and serum testosterone in US adult males.

MATERIALS AND METHODS

This was a cross-sectional study based on data from the National Health and Nutrition Examination Survey (NHANES) 2013-2014 and 2015-2016. The TyG index was calculated from fasting plasma glucose and serum triglyceride, and serum testosterone was measured by isotope dilution liquid chromatography tandem mass spectrometry in NHANES.

RESULTS

A total of 2186 male participants aged more than 20 years old were included, and the mean TyG index was 8.72 ± 0.71. Participants with a higher TyG index showed a lower level of total testosterone (β = -45.83, 95% CI: -58.50, -33.15, p < 0.0001) and a higher risk of testosterone deficiency (OR = 1.80, 95% CI: 1.46, 2.21, p < 0.0001) after we adjusted for all potential cofounders. Males in TyG index tertile 3 had a 113% (95% CI: 1.50, 3.02, p < 0.0001) higher risk of testosterone deficiency than those in tertile 1. Subgroup analysis stratified by diabetes condition indicated that the TyG index might be a good predictor of testosterone decline or deficiency in either people with or without diabetes. However, ROC cure analysis indicated that a larger area under the curve was found in the homeostasis model assessment of insulin resistance (HOMA-IR) (0.71, 95% CI: 0.68,0.74) than in the TyG index (0.67, 95% CI: 0.64, 0.70).

CONCLUSIONS

Males with a higher TyG index tended to have a higher risk of testosterone decline or even testosterone deficiency. However, the predictability of the TyG index for testosterone deficiency was not better than that of HOMA-IR. More well-designed studies are still needed to validate this relationship.