Abstract
Gestational diabetes mellitus (GDM) is conventionally diagnosed at 24–28 weeks of gestation, although hyperglycaemia may be detected earlier. Early gestational diabetes mellitus (eGDM), defined here as hyperglycaemia diagnosed before 24 weeks in the absence of overt diabetes, is characterised by limited but emerging molecular and metabolic evidence. In women with eGDM diagnosed before 20 weeks, DALI phenotyping identified insulin-resistant, secretion-dominant and mixed phenotypes using HOMA-IR and the OGTT-derived Stumvoll first-phase index, with greater obstetric risk in the resistance-dominant group. In the TOBOGM cohort, eGDM was associated with higher diacylglycerol and fatty-acid species, which remained associated with eGDM and glycaemic indices after adjustment for clinical risk factors. Early-pregnancy SHBG provides additional evidence, with a further reduction reported in women diagnosed before 20 weeks. The principal limitation is that eGDM-specific evidence remains sparse. Most early-pregnancy biomarker studies measured women before 24 weeks but diagnosed GDM later and therefore describe antecedents of subsequent disease rather than eGDM diagnosed before 24 weeks. Most mechanistic evidence also derives from conventional-window GDM or placental tissue obtained at delivery, preventing determination of whether reported abnormalities precede, accompany or result from early hyperglycaemia. Longitudinal studies with early diagnosis, concurrent assessment of insulin sensitivity and secretion, adjustment for adiposity and independent validation are therefore needed to define the molecular phenotype of eGDM.