The study aims to assess the utility of the triglyceride-glucose index (TyG) as a marker of insulin resistance (IR) in neonates. TyG and the homeostatic model assessment (HOMA-IR) values were compared in 196 singleton, term normoweight and without distress newborns. A Decision Tree procedure (CHAID) was used to classify cases into groups or predict values of a dependent (Ln HOMA-IR) variable. Three nodes were drawn for TyG: ≤ 6.7, > 6.7-7.8 and > 7.8 (p < 0.0001; F = 20.52). The predictability of those TyG values vs HOMA-IR was statistically significant (p < 0.0001). It was neither affected by gender (p = 0.084), glucose challenge test (p = 0.138) classifications nor by the TyG node* glucose challenge test and TyG node*gender interactions (p = 0.456 and p = 0.209, respectively). Glucose, HOMA-IR, and the triglyceride/HDL cholesterol ratio increased progressively from node 1 to 3 for TyG while QUICKI decreased.
Conclusion: In conclusion, TyG appears to be a suitable tool for identifying IR at birth, justifying the further insulin determination in those neonates. TyG ≥ 7.8 is recommended as cut-off point in neonates. The need for a follow-up study to confirm the TyG as early IR marker is desirable.
What is known: • HOMA-IR and the triglyceride-glucose index (TyG) show a high correlation. • The TyG has been used as an insulin resistance marker in adults.
What is new: • This is the first study where TyG has been assessed in neonates. • TyG appears to be a suitable and cheap tool for identifying insulin resistance at birth.
Keywords: Glucose; HOMA-IR; Insulin; Neonates; QUICKI; TG/HDLc; TyG.