Background and objectivesInsulin resistance (IR) is closely related to the decline or deficiency of testosterone in males. Triglyceride glucose-body mass (TyG-BMI) is considered to be a novel indicator of IR. We conducted this analysis to investigate the association between TyG-BMI and male testosterone, and to explore whether its ability to predict testosterone deficiency is superior to HOMA-IR and TyG.MethodsThis was a cross-sectional study using data from the National Health and Nutrition Examination Survey (NHANES, 2011–2016). The TyG-BMI index was calculated from serum triglyceride, fasting plasma glucose and BMI. The association of TyG-BMI with male testosterone was estimated by weighted multivariable regression.ResultsWe included 3394 participants for the final analysis. After adjusting for confounders, TyG-BMI was found to show an independent negative association with testosterone (β=-1.12, 95%CI: -1.50, -0.75, P<0.0001). Multivariate-adjusted beta also showed testosterone levels were significantly lower in the two highest TyG-BMI group (Q3, Q4) compared to the lowest group (Q1). Similar results were seen in all of the subgroup populations by stratified analysis (all P-interaction >0.05). Furthermore, ROC curve analysis indicated that the area under the curve of TyG-BMI index (0.73, 95% CI: 0.71, 0.75) was larger than that of HOMA-IR index (0.71, 95% CI: 0.69, 0.73) and TyG index (0.66, 95% CI: 0.64, 0.68).ConclusionOur result suggested a negative association between TyG-BMI index and testosterone in adult males. The predictability of the TyG-BMI index for testosterone deficiency is better than that of HOMA-IR index and TyG index.