Metabolic score for insulin resistance (METS-IR) as a predictor of gestational diabetes: Findings from a prospective Iranian cohort study

Abstract

Background

Gestational diabetes mellitus (GDM) represents a significant metabolic challenge in pregnancy, with relevance for Middle Eastern populations showing high prevalence rates. The metabolic score for insulin resistance (METS-IR) has emerged as a potential early predictor, though its performance in Iranian populations remains underexplored.

Methods

In this prospective cohort study conducted at Arash Women’s Hospital, Tehran (2020–2023), we enrolled 1,845 pregnant women during their first trimester (<12 weeks gestation). Participants underwent comprehensive metabolic evaluation including calculation of METS-IR. GDM diagnosis was established at 24–28 weeks using 75-g oral glucose tolerance test per American Diabetes Association criteria. We employed modified Poisson regression with robust variance estimation to assess associations, adjusting for key confounders identified through directed acyclic graphs.

Results

The study population demonstrated a GDM incidence of 19.78 % (365/1845). METS-IR showed strong predictive capacity with area under the ROC curve (AUC) of 0.82 (95 % CI: 0.79, 0.83). At the optimal cutoff of 2.36 (95 % CI: 2.30–2.41), sensitivity reached 72 % (95 % CI: 67, 76 %) and specificity 76 % (95 % CI: 74, 78 %). Adjusted risk ratios revealed a striking dose–response relationship across quartiles: Q2 = 4.21 (2.16, 8.23), Q3 = 8.85 (4.66, 16.78), and Q4 = 18.94 (10.16, 35.29) compared to the reference quartile.

Conclusion

This study establishes METS-IR as a robust early predictor of GDM in Iranian women, demonstrating superior performance to conventional metabolic markers. The simple calculation and strong predictive validity suggest its potential for first-trimester risk stratification in clinical practice, particularly valuable in high-prevalence populations.