Women with pre-gestational diabetes and their fetuses are at increased risk of developing serious complications compared with the non-diabetic pregnant women, includeing spontaneous abortion, preterm labor, hypertensive disorders, and delivery by cesarean section despite improved access and quality of antenatal care [1]. Women diagnosed to have gestational diabetes mellitus (GDM) during pregnancy also do develop similar complications, though not of same magnitude. Gestational diabetes mellitus accounts for ~90% of cases where pregnancy is complicated by diabetes [2], with potentially long-reaching consequences, including increasing the risk of subsequently developing type 2 diabetes for both mother and child [3]. Disrupted metabolic homeostasis has been implicated as a possible cause [4]; indeed, fetal macrosomia may result from minor disturbances in glucose metabolism [5]. These observations indicate an optimum glycemic control is essential to minimize the maternal and fetal morbidity and mortality of pregnancies complicated by glucose intolerance [6]. This is possible by home glucose monitoring by the glucometers which gives immediate glycemic level and by estimation of glycated proteins which indicate the glycemic excursions of the past few weeks to months.

Glycated proteins GA and A1C are known to reflect the plasma glucose level and are used as a standard index of glycemic control in patients with diabetes mellitus [7]. Since the lifespan of erythrocytes is about 120 days, A1C reflects the glycemia for the recent few months. Glycated albumin (GA) is an another index of glycemic control which correlates with the plasma glucose levels during the recent past few weeks because the turnover of albumin is about 20 days. Although the ratio of GA/A1C is usually close to 3, the value changes based on the metabolic state [8]. Several recent studies have confirmed that the point measurements of GA and A1C are closely correlated in diabetes. But publications are not many on the clinical use of GA as a marker for shortterm glycemic control in GDM. Hence this prospective study was undertaken with the aim of comparing GA with A1C and out of them which would serve as a earlier marker of glycemic control during pregnancy.

This study was initiated with the approval of ethics committee of the institute. A total of 524 pregnant women between 24 - 28 gestational weeks underwent 2 hrs OGTT with 75 g glucose from 21st Oct 2010 to 10th May 2012, in our “diabetes and pregnancy” specialty referral clinic. Among them, 271 were diagnosed as GDM by WHO criteria of 2 hr PG ≥ 7.8 mmol/L [9]. They were all explained about the study, particularly about the importance of adhering to the follow up schedule. A cohort of 30 GDM and 30 NGT women were included in this study after obtaining their informed consent. Details regarding family history, previous obstetric history, treatment for any concomitant diseases and food habits were obtained. Mean age, mean Body Mass Index (BMI) and mean Gestational Week (GW) were recorded for both GDM and NGT.

GDM were advised Medical Nutrition Therapy (MNT). The target glycemic control aimed was FPG ~ 5 mmol/L and 2 hr PPG ~ 6.6 mmol/L. Non-responders to MNT were administered insulin GDM were followed up every 2 weeks till 36 - 37 weeks. Venous blood samples were collected from them on each visit and their FPG, PPG were measured by GOD-POD method by Hitachi autoanalyzer 902. GA was estimated by Lucica GA-L assay kit—enzymatic method. For estimation of A1C the venous blood was collected at the first visit, third visit (4th week) and fifth visit (8th week). A1c was measured by automated determination with high-performance liquid chromatography (Bio-Rad). Ferritin level was measured by Ferro zinc electro chemiluminescence method—automated determination with cobas during first and fourth visits in both GDM and NGT. NGT were reviewed monthly once and the same biochemical parameters were determined.

Statistical analysis was performed by SPSS version 10. Comparison of mean values between the two groups was done by independent t-test and p value < 0.05 was considered statistically significant.

Mean age of GDM and NGT were 27.10 ± 4.05 and 23.80 ± 3.61 years; mean BMI was 26.50 ± 4.28 kg/m² and 22.10 ± 5.58 kg/m², mean GW was 26.23 ± 2.24 and 24.60 ± 1.87, respectively.

In GDM, mean FPG at first and last visit was 5.16 ± 0.55 mmol/L and 4.73 ± 0.52 mmol/L, respectively. PPG at first visit was 7.07 ± 1.51 mmol/L and 6.16 ± 0.70 mmol/L in the last visit, The mean GA was 12.48% ± 0.8%, 12.51% ± 0.9%, 12.40% ± 1.0%, 12.30% ± 0.86% and 12.38% ± 0.87% in first, second, third, fourth and fifth visit. The mean A1c at first  visit was 5.16% ± 0.35%, in the third visit (4th week) 5.24% ± 0.29% and in the last visit (8th week) 5.216% ± 0.28%. In this study both GA and A1C values run parallel (Figure 1). In NGT women, mean FPG in the first was 4.37 ± 0.37 mmol/L and in the last visit 4.39 ± 0.43 mmol/L. Mean PPG was 5.95 ± 1.01 mmol/L in the first visit and 5.75 ± 1.61 mmol/L in the last visit. Mean GA was 12.17% ± 0.85% and 12.10% ± 0.77% at the first and the last visit respectively. A1c was 4.84% ± 0.31% and 4.91% ± 0.33% in the first and the last visit respectively. Ferritin value for GDM was 26.2 ± 11.4 ng/mL and 26.7 ± 12.5 ng/mL in first & fifth visits. The glycemic control was observed earlier with GA, around two weeks compared to eight weeks by A1C. Anthropometric findings and biochemical parameters are given Table 1 for GDM and Table 2 for NGT groups. Among 30 GDM, 7 responded to MNT and the remaining required insulin for glycemic control.

Obesity and advanced age are the risk factors for the

Table 1. Biochemical parameters for GDM group mean values.

Table 2. Biochemical parameters for NGT Group Mean Values.

development of GDM [10]. In this study also, these factors were observed as risk factors for GDM. During pregnancy a strict glycemic control is essential to prevent the perinatal complications. The ideal glycemic control for this would be FPG ~ 5 mmol/L, 2 hr PPG ~ 6.6 mmol/L and mean plasma glucose 5.85 mmol/L [11]. The therapy is adjusted by monitoring the plasma glucose both in the fasting and post prandial state. To understand the glycemic excursions during pregnancy besides performing the point of care by capillary blood glucose measurement with glucometers, there is a need to have a test which could give an idea about the plasma glucose excursions in the previous weeks. GA as well as A1C are useful markers of glycemic control in the recent past [12]. GA reflects the retrospective glycemic excursions in the immediate past of approximately two weeks and A1C, glycemic excursions of more than eight weeks.

There is a biphasic variation of A1C levels during pregnancy [13]. The biphasic variation of A1C levels in subjects with NGT [14] and GDM [15] is explained by iron-deficiency anemia in them. This phenomenon is attributed to increase in A1C as a result of extension of red cell survival due to iron deficiency anemia [16-18] and these findings have been confirmed by Kim C. et al. [19]. GA is not affected by this phenomenon and thus serves as a better indicator of short term glycemic control.

In this study, GA levels strongly correlated with the A1C levels in both GDM and NGT women and these glycated proteins also correlated with FPG and PPG, similar to the observation of Yang et al. and Abe et al. [10,12]. Hashimoto K et al. [14] documented that A1C but not GA was elevated because of iron deficiency anemia in late pregnancy in diabetic women. In our cohort, we did not find this difference as the ferritin level was within the normal range. This is due to the public health policy of iron and folic acid supplementation from the preconception period itself. In this study both GA and A1C values run parallel. This is due to the intervention with MNT or insulin to maintain the maternal plasma glucose at the target glycemic level. Another observation in this study is GA indicates maternal glycemic control of the past few weeks whereas A1C level indicates of the past few weeks to months, but the important clinical advantage of GA is that, it reveals glycemic excursions earlier.

GA is a better indicator of short term glycemic excursions than A1C. GA offers an opportunity for earlier interventions to obtain a better glycemic control during pregnancy and thus good fetal outcome. We recommend more studies in this aspect.

We thank the patients for their co-operation during the study. This study received financial support from Asahi Kasei Pharma, Japan.