Factors predicting the need for insulin therapy in patients with gestational diabetes mellitus

Abstract

Aim

To determine the possible factors predicting the insulin requirement in pregnancies complicated by gestational diabetes mellitus (GDM).

Method

A total of 294 patients with GDM diagnosed by the 100-g/3-h oral glucose tolerance test (OGTT) were studied. The following factors were analyzed: maternal age, nulliparity, family history of diabetes, prepregnancy BMI, prior GDM, prior fetal macrosomia, multiple pregnancy, polyhydramnios, gestational age at diagnosis of GDM, smoking, hypertension, number of abnormal 100-g/3-h OGTT values, and glycated hemoglobin (HbA1c). The association between each factor and the need for insulin therapy was then analyzed individually. The performance of these factors to predict the probability of insulin therapy was estimated using a logistic regression model.

Results

Univariate analysis showed a positive correlation between insulin therapy and prepregnancy BMI, family history of diabetes, hypertension, prior GDM, prior fetal macrosomia, number of abnormal 100-g/3-h OGTT values, and HbA1c (P < 0.05). Prepregnancy BMI, family history of diabetes, number of abnormal 100-g/3-h OGTT values and HbA1c were statistically significant variables in the logistic regression model.

Conclusions

The probability of insulin therapy can be estimated in pregnant women with GDM based on prepregnancy BMI, family history of diabetes, number of abnormal 100-g/3-h OGTT values, and HbA1c concentration.

Introduction

Gestational diabetes mellitus (GDM) is currently defined as any degree of carbohydrate intolerance with onset or first recognition during pregnancy. This definition, proposed by the American Diabetes Association (ADA) in 1998 [1], has been widely used because it reliably contemplates the physiopathology of the disease in a simple and objective manner. In view of the importance of this disease considering both its high prevalence and severe repercussions if adequate treatment is not instituted, the diagnosis of the largest possible number of patients at risk of GDM justifies this broad definition which is being used worldwide.

On the other hand, it should be emphasized that the definition of GMD comprises a set of diagnostic possibilities and different degrees of glucose intolerance. This broad spectrum of disease severity seems to be associated with perinatal outcomes [2], [3] and with the risk of developing type 2 diabetes during the adult life of these women [4]. It seems possible to weigh these different degrees of glucose intolerance during pregnancy in terms of severity, but it remains uncertain above which threshold the risks for the pregnant woman and especially for the fetus reach a magnitude that requires more intensive and/or differentiated monitoring [5].

The severity of GDM is associated with maternal glucose levels that present a positive and direct correlation with the risk of fetal involvement [2], [5], [6], [7], [8], [9]. In the case of women who do not achieve desired glucose levels with diet and exercises, drug therapy should be instituted to reduce glucose levels in order to guarantee good fetal development and to minimize neonatal complications [8], [9]. The only drug approved so far by the Food and Drug Administration (FDA) for the use in diabetic pregnant women is insulin [10], [11]. Thus, the need for insulin therapy might be a starting point for the characterization of patients with more severe GDM due to a greater difficulty in glycemic control [12]. The search for factors effectively predicting which patients will potentially require insulin is a way to grade glucose intolerance and to direct attention and resources to a subpopulation of patients with more severe GDM.

Some risk factors that indicate the introduction of insulin therapy for glycemic control in pregnancies complicated by GMD have been studied. It is believed that a diagnosis of GMD at an early gestational age or based on fasting glucose levels, as well as the presence of obesity or a family history of diabetes or exacerbated fetal growth, are associated with a more severe degree of glucose intolerance [3], [13], [14], [15], [16]. Similarly, the interpretation of 100, 75 or 50 g oral glucose tolerance tests indicates the possible existence of a directly proportional relationship between the severity of the disease and the number of abnormal blood glucose values (above the reference limits) or the magnitude of glycemic alteration for each value analyzed [17], [18]. The role of glycated hemoglobin in GDM remains controversial despite attempts to correlate this parameter with perinatal outcomes [19]. None of these studies, however, correlates the degree of glucose intolerance with the use of insulin.

The objective of the present study was to determine the association between clinical and laboratory parameters and insulin requirement in pregnancies complicated by GDM, and to evaluate possible factors predicting insulin need. These predictive factors were then used to estimate the probability of the need for insulin therapy in these patients and to identify subgroups with different degrees of glucose intolerance.