The Bernese Gestational Diabetes (GDM) Project: Third trimester fasting glucose or first trimester HbA1c to reduce oral glucose tolerance tests?

Introduction According to swiss guidelines using a fasting glucose (FG) value of < 4.4 mmol/l as lowest threshold we could withhold an oGTT. Aim of this study was to investigate the negative predictive value (NPV) of a third trimester FG < 4.4mmol/l to exclude a GDM.

Material and methods We performed a prospective study on low and high risk women for GDM. All had an oGTT performed between 24 and 28 weeks of gestation and an HbA1c value in the first trimester. Women with an HbA1c value ≥ 6.5 % were excluded. The performance of a FG cut-off of 4.4mmol/l was calculated. ROC curve analysis was used to explore the screening capability of FG in GDM diagnosis. Statistical significance was considered when p-value < 0.05.

Results We included 735 women. The incidence of GDM was 14.8 %. Based on the FG< 4.4mmol/l, we would have missed 16 (5 %) GDM cases. Of interest, 3/16 (18 %) needed Insulin during pregnancy and showed polyhydramnios. One women had even a pathological postpartum oGTT. A value below < 4.40 mmol/l as proposed to exclude GDM showed a sensitivity of 47 %, specificity of 89 %, and a NPV of 0.64. In comparison the NPV of HbA1c < 4.5 % in the first trimester is 0.99.

Conclusion We showed that a HbA1c < 4.5 % performs better than a third trimester FG < 4.4mmol/l in excluding GDM. Using FG< 4.4mmol/l, we would have missed a small percentage of GDM, however including women with metabolic disorder diagnosed after birth or needing insulin therapy.

Publication History

Article published online:
07 October 2020