Preconception 1-hour blood glucose as an early indicator of gestational diabetes mellitus risk in PMOS/PCOS women undergoing IVF treatment: a multi-centre cohort study

STUDY QUESTION Is the preconception 1-h plasma glucose level during the oral glucose tolerance test (OGTT) more strongly associated with subsequent gestational diabetes mellitus than the fasting or 2-h glucose levels in women with PMOS/PCOS undergoing IVF treatment? SUMMARY ANSWER Preconception 1-h plasma glucose showed a stronger and more consistent association with gestational diabetes mellitus than the fasting or 2-h glucose levels in women with PMOS/PCOS undergoing IVF treatment. WHAT IS KNOWN ALREADY Women with PMOS/PCOS have an increased risk of gestational diabetes mellitus. International guidelines recommend a preconception OGTT in women with PMOS/PCOS without pre-existing diabetes because of the risk of hyperglycaemia and pregnancy complications. However, the comparative predictive value of different preconception OGTT time points, particularly 1-h glucose, remains unclear. STUDY DESIGN, SIZE, DURATION This was a multi-centre retrospective cohort study including 2571 women with PMOS/PCOS who underwent their first IVF treatment at three reproductive medicine centres in China and had a preconception 75-g OGTT between March 2018 and December 2023. PARTICIPANTS/MATERIALS, SETTING, METHODS Women with PMOS/PCOS who completed a preconception 75-g OGTT before initiation of the IVF treatment were included. The primary outcome was GDM diagnosed at 24–28 weeks of gestation. Generalized linear mixed-effects models were used to assess associations between OGTT glucose levels and GDM, adjusting for age, BMI, infertility duration, hormonal profiles, and treatment protocols. Stratified analyses by BMI and PMOS/PCOS phenotype were conducted, and multiple imputation was applied for missing data. MAIN RESULTS AND THE ROLE OF CHANCE Preconception fasting (aOR 1.97 [1.49–2.61]), 1-h (aOR 1.23 [1.13–1.34]) and 2-h glucose (aOR 1.30 [1.17–1.45]) were all significantly associated with GDM (all P < 0.001). Within the diabetic range, only the 1-h glucose measurement of ≥11.6 mmol/l remained a significant indicator. The association of 1-h glucose with GDM was significant in normal-weight and obese women but not in overweight women. The associations were consistent across PMOS/PCOS phenotypes. LIMITATIONS, REASONS FOR CAUTION The retrospective design may have introduced residual confounding and limits causal inference. Missing data required imputation, and subgroup analyses may have limited statistical power. WIDER IMPLICATIONS OF THE FINDINGS These findings suggest that preconception 1-h glucose may provide additional value for identifying women who are a higher risk of GDM among normal-weight and obese women with PMOS/PCOS undergoing IVF treatment, and support its consideration in preconception metabolic assessments. FUNDING This work was supported by the National Natural Science Foundation of China (82522038, 82471670 and 82401908), the Natural Science Foundation of Fujian Province (2025J011428), and the “Yongjiang Science and Technology Innovation 2035” Key Research and Development Program (2025Z151). DISCLOSURES The authors declared no competing interests. TRIAL REGISTRATION NUMBER N/A