Postpartum Glycaemic Phenotype and Continuous Glucose Monitoring at 4-6 Years After Gestational Diabetes Defined by Different Diagnostic Criteria.

AIM: To characterise glycaemic outcomes 4-6 years postpartum according to gestational diabetes mellitus (GDM) diagnostic thresholds and evaluate whether continuous glucose monitoring (CGM) identifies differences not captured by conventional clinical measures. METHODS: Women recruited before and after introduction of WHO GDM criteria were followed at 4 and 6 years postpartum. At each follow-up, participants underwent oral glucose tolerance testing (OGTT), HbA1c measurement, body composition assessment and 14-day CGM. Based on pregnancy OGTT results and the diagnostic criteria in use at the time of testing, women were classified as meeting EASD 1991 criteria (GDMOLD; higher glucose thresholds), meeting WHO 2013 criteria but not EASD 1991 without diagnosis and treatment (GDMWHO-) or meeting WHO 2013 criteria with diagnosis and treatment (GDMWHO+). RESULTS: Women classified as GDMOLD had higher fasting glucose at both follow-ups compared with women meeting only WHO 2013 criteria, and showed greater hyperglycaemic exposure and glycaemic variability on CGM. Time above range, coefficient of variation and mean amplitude of glycaemic excursions were consistently higher in GDMOLD group, and differences remained significant after adjustment for pregnancy glucose levels. Within the WHO-defined groups, conventional clinical measures did not differ between women with and without a GDM diagnosis, whereas CGM detected modest differences in mean glucose and time above range over time. Approximately 50% of women meeting only WHO 2013 criteria had prediabetes. CGM metrics were strongly associated with OGTT-derived glucose and HbA1c but only weakly with anthropometric and blood lipid measures. BMI and body fat percentage did not differ between groups. CONCLUSIONS: Long-term glycaemic risk aligns with diagnostic thresholds used during pregnancy. Women diagnosed by older, higher thresholds exhibit a persistently less favourable glycaemic profile. Women meeting the lower WHO 2013 glucose thresholds represent a large group with a high prevalence of prediabetes, regardless of diagnosis during pregnancy. CGM detects modest differences in free-living glycaemic patterns not apparent from conventional clinical measures.