Apotential etiology for gestational diabetes mellitus (GDM) is a limitation in β-cell reserves that manifests as hyperglycemia when insulin secretion does not increase to match the escalated insulin needs during pregnancy (1). Recently, the international association of diabetes and pregnancy study groups (IAPDSG) proposed new criteria for diagnosing and classifying diabetes in pregnancy based on the association of maternal hyperglycemia with perinatal outcomes (2). To date, β-cell function in GDM defined by the new consensus criteria has not been reported. β-cell function has been assessed by a disposition index defined as the product of insulin sensitivity and insulin secretion determined by intravenous glucose tolerance test (IVGTT) (3). The recent demonstration of a hyperbolic relation between indexes of insulin secretion and insulin sensitivity from the oral glucose tolerance test (OGTT) has led to the introduction of novel OGTT-derived measures of β-cell function analogous to the disposition index (4). Therefore, we assessed β-cell function in Japanese women with GDM using the OGTT-derived measures (i.e., oral disposition index [DIo]). Of the 2,267 sequential women who underwent universal screening for GDM from January 2004 to February 2010, 409 with positive 1-h 50-g glucose challenge test results had a standard 2-h 75-g OGTT. Venous blood samples for the measurement of plasma glucose level (mg/dl) and insulin concentration (μU/ml) were drawn in the fasting state, at 30 min, 1 h, and 2 h after ingestion of the glucose drink. The ratio of the insulin area under the curve to the glucose area under the curve multiplied by the composite index of insulin sensitivity during the OGTT was used as the DIo in this study. With the IAPDSG criteria, 126 women were diagnosed as GDM: 58 with single abnormal value, 50 with two abnormal values, and 18 with three abnormal values. Of the 126 GDM women, 101 (80%) were identified by the fasting glucose plus 1-h plasma glucose levels. The natural log-transformed DIo in GDM was significantly lower than that in non-GDM (mean ± SD, 1.69 ± 0.57 vs. 2.59 ± 0.96, P < 0.0001). Similar to women with two or three abnormal OGTT values (1.62 ± 0.35 and 1.06 ± 0.38, respectively), those with a single abnormal value exhibited significantly lower levels of DIo (1.96 ± 0.59) compared with those with non-GDM (P < 0.001). After adjustment for prepregnancy BMI, family history of diabetes, and the glucose intolerance status (i.e., GDM) using multiple linear regression model, GDM was still a negative correlate of the DIo (P < 0.0001).
Consistent with the IADPSG analysis, fasting plus 1-h plasma glucose levels identified the majority of GDM women. Using the DIo, β-cell dysfunction was demonstrated in women with GDM defined by the IAPDSG criteria. Previous studies indicate that β-cell function determined by the disposition index from OGTT as well as IVGTT is predictive of the development of diabetes (5). Our findings suggest that the women with GDM identified using the new consensus criteria are at high risk of development of subsequent diabetes, as well as adverse perinatal outcomes.
Acknowledgments
No potential conflicts of interest relevant to this article were reported.
The authors are grateful to Drs. Keisuke Kouyama and Takayuki Abe of Center for Clinical Research, Keio University School of Medicine, for their valuable statistical assistance.
