Association of First-Trimester Aneuploidy Markers with Gestational Diabetes Mellitus

Abstract

Introduction: The earliest identification of women who are at risk of developing GDM is becoming increasingly important to improve the prognosis, as the prevalence of gestational diabetes mellitus is rising. Our goal is to determine whether first-trimester aneuploidy markers are associated with the eventual development of gestational diabetes mellitus.

Objectives: To investigate any association of first-trimester aneuploidy markers nuchal translucency (NT), pregnancy-associated plasma protein-A (PAPP-A), human chorionic gonadotropin (β-hCG), and ‘Combined risk’ (ultrasound and biochemical marker) with gestational diabetes mellitus (GDM). 

Material and methods: In this prospective cohort study, first-trimester aneuploidy marker tests were performed between 11-13+6 weeks of gestation and GDM screening was done at the first visit, 24-28 weeks, and 32-34 weeks of gestation. The antenatal women were divided into GDM and non-GDM as per IADPSG glycemic cut-offs. Demographic, clinical, and laboratory data of both groups were compared and predictive tests were applied to detect GDM.

Results: A total of 292 patients, 70 (23.9%) in the GDM group and 222 (76%) in the non-GDM group were included in the study. The median age, crown-rump length and nuchal translucency data were statistically similar in both groups. PAPP-A MoM (GDM:1.02 ± 0.70; non-GDM:0.95 ± 0.56; p=0.535) and β-hCG MoM (GDM: 1.04 ± 0.45; non-GDM: 1.07 ± 0.57; p=0.092) levels of the GDM group were similar to non-GDM group statistically. β-hCG >1.5 MoM (p=0.0018), predicts a lower risk of GDM development. On ROC analysis, β-hCG MoM had a higher area under the curve than PAPP-A.  β-hCG at a cut-off of 1.21 MoM, the sensitivity of detecting GDM was 80% and the specificity was 37%, whereas β-hCG at 1.08 MoM, the sensitivity was 60% and the specificity was 50%.

Conclusions: Among the first-trimester aneuploidy markers, only β-hCG MoM can be utilized as a screening test to predict GDM with low predictability.