Discordant muscle-adipose phenotypes and insulin resistance in prediabetes: sex-specific associations and predictive utility of TyG-WHtR combined with HDL-C

Objective We aimed to explore the interplay between muscle-adipose composite phenotypes and insulin resistance (IR) among individuals with prediabetes(Pre-DM), and to evaluate the discriminative capacity and sex-specific variations of the triglyceride-glucose-waist-to-height ratio (TyG-WHtR) when combined with high-density lipoprotein cholesterol (HDL-C). Methods A total of 167 subjects with Pre-DM were enrolled and categorized into four phenotypes—high-muscle/low-fat, high-muscle/high-fat, low-muscle/high-fat, and low-muscle/low-fat—based on their appendicular skeletal muscle index and visceral fat level. Fasting blood glucose, lipid profiles, and body composition parameters were measured, and the TyG index and its derived indices were calculated. Hierarchical linear regression and logistic regression were used to evaluate the independent and interactive contributions of these phenotypes, lipid profiles, and TyG-derived indices to HOMA-IR. Subsequently, a nomogram was constructed using binary logistic regression, and its performance was evaluated via the area under the receiver operating characteristic curve (AUC), calibration curve, decision curve analysis (DCA), and clinical impact curve (CIC). Results Significant intergroup differences emerged for body mass index, waist circumference, fasting insulin, HOMA-IR, HDL-C, and TyG-derived indices (all P < 0.05). Spearman’s correlation analysis revealed that fasting insulin and HOMA-IR were positively correlated with WHtR, waist circumference, BMI, body fat mass, visceral fat level, body fat percentage, and ASMI (all P < 0.01). Triglycerides showed a positive correlation with these metrics, whereas HDL-C exhibited a negative correlation (all P < 0.05). Notably, TyG-WHtR displayed the strongest correlation with HOMA-IR ( P < 0.001). Hierarchical regression identified HM/HF as the strongest independent predictor of HOMA-IR, remaining significant after full adjustment (B = 3.264, P < 0.001). The protective effect of LM/LF was attenuated to marginal significance after adjusting for blood lipids (B=−0.705, P = 0.082), suggesting that this association largely reflects low visceral fat rather than muscle mass per se. The final model accounted for 36.0% of the variance in HOMA-IR. We did not detect a statistically significant multiplicative interaction ( P =0.089); however, the study may be underpowered, and additive interaction was not formally assessed. In exploratory stratified analyses, the HM/HF effect was more pronounced in men (B = 5.662) and in adults aged <60 years (B = 4.425), whereas the protective LM/LF effect was restricted to women (B=−0.914). In the ≥60-year group, the sex coefficient changed from negative to positive without reaching significance, suggesting that age may modify the sex effect; formal interaction testing is needed to confirm this possibility. Logistic regression identified TyG-WHtR (OR=8.231) and HDL-C (OR=0.183) as independent determinants of IR; the corresponding Cohen’s d values exceeded the conventional large-effect threshold of 0.8. The combined nomogram achieved an AUC of 0.815 (95% CI 0.748–0.881). The Hosmer-Lemeshow test showed no evidence of departure from perfect calibration, and the calibration curve derived from 1000 bootstrap resamples closely followed the ideal diagonal line. Decision curve analysis indicated that, across threshold probabilities of 0.2–0.8, the net benefit remained consistently greater than that of both the “intervene-all” and “intervene-none” strategies. Conclusion In Pre-DM, insulin resistance appears to be driven primarily by visceral adiposity rather than overall obesity. The metabolic impact of skeletal muscle is highly dependent on the surrounding lipid milieu. The interaction term between low muscle mass and high visceral fat was non-significant, with no multiplicative interaction detected. Consequently, it remained inconclusive whether their effects on IR followed an additive pattern. Sex and age may modify the association between body composition and IR, although formal interaction testing is needed to confirm this. The nomogram integrating TyG-WHtR and HDL-C demonstrated good calibration and may serve as a practical screening tool in primary care settings. Maintaining moderate muscle mass in the context of low visceral adiposity is likely to confer greater metabolic benefit.