Abstract Aims To estimate the prevalence of gestational diabetes mellitus (GDM) in a Danish cohort comparing the current Danish versus the WHO2013 diagnostic criteria, and to evaluate adverse pregnancy outcomes among currently untreated women in the gap between the diagnostic thresholds. Methods Diagnostic testing was performed by a 75 g oral glucose tolerance test (OGTT) at 24–28 weeks’ gestation in a cohort of pregnant women. GDM diagnosis was based on the current Danish criterion (2-h glucose ≥ 9.0 mmol/L, GDM_DK) and on the WHO2013 criteria (fasting ≥ 5.1, 1 h ≥ 10.0 or 2 h glucose ≥ 8.5 mmol/L, GDM_WHO2013). Currently untreated women fulfilling the WHO2013 but not the Danish diagnostic criteria were defined as New-GDM-women (GDM_WHO2013-positive and GDM_DK-negative). Adverse outcomes risks were calculated using logistic regression. Results OGTT was completed by 465 women at a median of 25.7 weeks’ gestation. GDM_DK prevalence was 2.2% (N = 10) and GDM_WHO2013 21.5% (N = 100). New-GDM was present in 19.4% (N = 90), of whom 90.0% had elevated fasting glucose. Pregnancies complicated by New-GDM had higher frequencies of pregnancy-induced hypertension (13.3% vs 4.1%, p = 0.002), large-for-gestational-age infants (22.2% vs 9.9%, p = 0.004), neonatal hypoglycaemia (8.9% vs 1.9%, p = 0.004) and neonatal intensive care unit admission (16.7% vs 5.8%, p = 0.002) compared to pregnancies without GDM. Conclusions GDM prevalence increased tenfold when applying WHO2013 criteria in a Danish population, mainly driven by higher fasting glucose levels. Untreated GDM in the gap between the current Danish and the WHO2013 diagnostic criteria resulted in higher risks of adverse pregnancy outcomes.