Association between glycated haemoglobin, glycated albumin and fructosamine with neonatal birthweight and large-for-date status infants in gestational diabetes mellitus: a prospective cohort study.

This study aims to investigate associations between glycated haemoglobin (HbA1c), glycated albumin (GA) and fructosamine with neonatal birthweight in gestational diabetes mellitus (GDM). The prospective cohort consisted of 82 women with GDM and their newborns, enrolled between November 2016 and September 2017. Considering neonatal birthweight and birthweights ≥90th percentile for gestational age as outcomes, linear and logistic regression models were used, respectively. Fructosamine (R2 =0.62) and GA (R2 =0.61) performed very similarly between them and best than HbA1c (R2 =0.58). The added value of GA or fructosamine to HbA1c resulted in increase in models' performances. GA attained the best discriminative ability regarding large-for-date status babies (AUC = 0.80, OR-estimate 1.58, p=.001) followed by fructosamine (AUC = 0.78, OR-estimate 1.42, p=.001) and HbA1c (AUC = 0.69, OR-estimate 3.09, p=.070). GA and fructosamine, besides from providing additional information to HbA1c, when used separately perform better than the traditional biomarker in predicting neonatal birthweight and large-for-date babies in pregnant women with GDM. Impact statement What is already known on this subject? HbA1c is the standard glycaemic indicator used in GDM. Its association with birthweight and large-for-date status has been previously reported. However, it has become increasingly questionable whether it is a suitable glycaemic marker in pregnancy. There is a growing interest in other non-traditional shorter-term glycaemic indicators, such as GA and fructosamine. Nevertheless, few studies exist and almost all are retrospective and with ethnically homogeneous study populations composed by pregnant women not only with GDM but also type 1 and type 2 diabetes mellitus. What do the results of this study add? Our prospective multi-ethnic cohort composed solely on pregnant women with GDM and their infants show that even though all of the aforementioned biomarkers are associated with birthweight and large-for-date status in GDM when used separately, GA and fructosamine seem to perform better than HbA1c. When used with HbA1c, they improve the predicting performance of the traditional marker. What are the implications of these findings for future clinical practice and/or further research? These findings suggest that GA and fructosamine can provide important additional or substitute information to HbA1c in GDM, namely in predicting birthweight and large-for-date status babies. Larger studies are needed to confirm if this non-traditional biomarkers can change clinical practice.