Gestational diabetes mellitus remains a riddle wrapped in a mystery inside an enigma

Abstract

Dear Sir, We appreciate the comments of Dr. Ozteikin about our paper on HbA1c as a screening test for gestational diabetes mellitus (GDM), which was recently published in this journal. Despite four decades of research, GDM remains, ‘a riddle wrapped in a mystery inside an enigma,’ without any consensus on the screening, diagnosis, and follow-up guidelines of the various expert panels (1). In high-risk populations like the one in our study, the American Diabetes Association and the American College of Obstetricians and Gynecologists recommend screening all pregnant women (i.e. universal screening) with the one-step oral glucose tolerance test (OGTT) at first booking, as early in the pregnancy as possible. All women testing negative need to repeat the OGTT between 24 and 28 weeks’ gestation to rule out GDM. Thus, every pregnant woman has the potential to undergo two OGTTs in the antenatal period. This approach, though conceptually sound, is very demanding and difficult to follow in clinical practice. We tend to follow a practical approach that is a hybrid of these guidelines. Our study was done in a clinical setting with the OGTT being scheduled between 24 and 28 weeks. However, many women were tested earlier due to a clinical decision, as they were at a much higher risk for GDM. Many pregnant women presented for the first time and for first booking anywhere between 28 and 40 weeks. This is the reason for the wide range of gestational ages (16–40 weeks) seen in our study. We repeated the analyses of our data using only women between gestational ages 24 and 28 weeks, who constituted 44% of all the pregnant women in our study. The area under the receiver operating characteristic curve of HbA1c to detect GDM was 0.50 (95% CI 0.38–0.62). This was no different than the AUC for the entire cohort (0.54) as stated in our paper, signifying again that HbA1c is a test of no value in screening for GDM. Thus, our basic message remains the same, i.e. despite all the progress in technology, HbA1c remains a poor test to screen for GDM.