Effects of antidiabetic agents on pancreatic beta-cell function in gestational diabetes: is there enough evidence?

ABSTRACT

Gestational diabetes mellitus (GDM) is typically characterized by the presence of insulin resistance. However, recent studies showed that both insulin resistance and pancreatic beta-cell function impairment may contribute to the development of type 2 diabetes in women with history of GDM. In fact, beta-cell function decline was found as significant predictor of later disease in former GDM women progressing towards type 2 diabetes. Despite the evidence of the relevance of beta-cell function quantification in GDM, a low number of studies focused on the effects of GDM treatments on beta-cell function. We briefly present the evidence of the effects on beta-cell function of pharmacological agents, as well as nutrition supplements or medical nutrition therapy, used in the management of GDM. We found that few studies reported information on beta-cell function effects in GDM, despite some agents, such as glyburide, are well known insulin secretagogues. Therefore, further studies should be carried out to clearly assess the effects on beta-cell function of the treatments in GDM women.

KEYWORDS:

• Antidiabetic agents
• beta-cell function
• gestational diabetes
• hypoglycemic agents
• insulin secretion
• nutrition therapy

1. Introduction

It is well known that gestational diabetes mellitus (GDM) is typically characterized by the presence of insulin resistance. In some studies, however, impairment in pancreatic beta-cell function was identified as a concomitant defect. Indeed, recent investigations showed that both insulin resistance and beta-cell function impairment may contribute to the development of type 2 diabetes in women with history of GDM.[1–4] The study by Tura et al. [1] suggested that the critical event for the onset of type 2 diabetes was a marked deterioration of insulin sensitivity, but also beta-cell function showed a progressive decline in the women developing the disease; in fact, in the progressors, beta-cell function impairment was found as significant predictor of the subsequent disease development.

Despite the evidence of the relevance of beta-cell function quantification in GDM, still it appears that more attention was devoted to insulin resistance. In fact, the effects of treatments on insulin sensitivity have been well studied in GDM. Conversely, few studies focused on the effects of treatments on beta-cell function.

2. Pharmacological treatment of GDM

Several studies [5,6] reported a comprehensive picture of the possible options for the pharmacological treatment of GDM. Insulin was the therapy of choice for glycemic control in women with GDM. In Ref. [5], it was reported that human and neutral protamine Hagedorn insulin do not cross the placenta, and can achieve a satisfactory glycemic control. As regards more recent short-acting and long-acting insulin, it is not completely excluded that they can cross the placenta, but at any rate they do not appear to determine adverse fetal effects.[5] Insulin, however, requires multiple-daily injections. Also, there is a significant risk for hypoglycemia. For these reasons, oral antidiabetic agents have become appealing: there is no need of self-injections, and they are associated to a lower hypoglycemic risk.

Glyburide and metformin are the most common oral agents used for the management of GDM. Currently, several other agents have become possibly available, such as non-sulfonylurea secretagogues (nateglinide, repaglinide), thiazolidinediones (pioglitazone, rosiglitazone), α-glucosidase inhibitors (acarbose, miglitol) and dipeptidyl peptidase IV inhibitors (sitaglipin and others), but for these agents, human data in pregnancy are still limited.[6]

Glyburide is a sulfonylurea, which increases insulin secretion and, possibly, insulin sensitivity in peripheral tissues. Glyburide decreases mean glucose by approximately 20%. However, not all women achieve optimal glycemic control with glyburide. Women who fail glyburide therapy are typically older, with higher fasting glycemia, and severely overweight.

Metformin is a biguanide, an insulin sensitizer. It stimulates glucose uptake in the liver and peripheral tissues. Also, it inhibits gluconeogenesis, suppresses hepatic glucose output and increases intestinal glucose absorption. Thus, metformin improves insulin sensitivity by reducing both fasting plasma glucose and insulin concentration. Metformin was not widely used in GDM until the occurrence of a crucial study, the MiG trial.[7] In that study, metformin was proven to be equivalent to insulin in terms of the study’s primary outcome, though, on the other hand, the failure rate of metformin to achieve optimal glycemic control, thus requiring insulin supplementation, was relatively high (46% of the patients).

3. Evidence of the effects of treatment on beta-cell function

Several controlled trials were devoted to the study of women with GDM under pharmacological treatment. However, to our knowledge, few studied reported quantitative information on possible amelioration of beta-cell function.[8–10]

The study by Hebert et al. [8] included 40 women with GDM, 40 healthy pregnant and 26 non-pregnant women with type 2 diabetes. All women with GDM, and the majority of women with type 2 diabetes, were treated with glyburide twice daily. All women underwent a mixed meal tolerance test, with measurement of plasma glucose, insulin and C-peptide. Some parameters of beta-cell function were estimated through appropriate mathematical modeling. It was found that total beta-cell responsivity (beta-cell ability to secrete insulin in response to glucose stimulation) was similar in GDM and healthy pregnant women, thus reflecting the pharmacological activity of glyburide. However, the effects on beta-cell function of glyburide were not enough to compensate for the degree of insulin resistance in GDM women, as reflected by the significantly lower disposition index (beta-cell response to glucose corrected for the prevailing insulin sensitivity).

In the study by Glueck et al.,[9] a group of 39 women with polycystic ovary syndrome, treated with metformin, were studied before and during pregnancy. Some of them developed GDM, but due to the limited number of cases, results were not presented separately for women with and without diabetes. According to HOMA beta-cell index, essentially based on the ratio of insulin to glucose at fasting condition only, it was found that in the pre-conception condition metformin decreased beta-cell function, but during pregnancy beta-cell function recovered, reaching the pre-conception values.

In the study by Xiang et al.,[10] 59 women, treated with pioglitazone or placebo (27 and 32 women, respectively) completed the study period, with baseline and post-treatment intravenous glucose tolerance test. It should be observed, however, that women were not studied during pregnancy, the study being focused on women with a previous history of GDM, but with non-diabetic condition at the time of the study. Results showed that there was no significant improvement in beta-cell function (disposition index) after 1 year of treatment. Of note, insulin sensitivity was not improved either.

Some studies reported possible effects on beta-cell function in the case of treatment with nutrients, rather than actual pharmacological agents.[11–13] In the study by Mozaffari-Khosravi et al.,[11] 45 postpartum women with previous GDM were divided into an intervention and a control group (24 and 21, respectively). Women in the intervention group received an injection of 300,000 IU of vitamin D. Three months after the intervention, they were revaluated, and it was found that beta-cell function (assessed by HOMA beta-cell index) was slightly increased only in the control group. On the other hand, when comparing beta-cell function in the two groups, it was neither different before nor after the intervention. In the study by Yeow et al.,[12] a double-blind, randomized controlled trial was performed on 26 women with a previous history of GDM, which were randomized to receive either daily 4,000 IU vitamin D or placebo for 6 months. Beta-cell function was evaluated at baseline and at the end of the treatment period through a 75-g oral glucose tolerance test (OGTT), with several empirical indices: the difference between one hour and fasting insulin normalized to the same difference in glucose (sort of insulinogenic index), the area under the curve (AUC) of insulin during the OGTT normalized to glucose AUC, a similar index with C-peptide, an OGTT-based acute insulin response index, and the oral disposition index (insulin to glucose AUC ratio, multiplied by an index of insulin sensitivity). In this study, almost all indices showed an improvement in beta-cell function due to vitamin D.

In the study by Samimi et al.,[13] 56 GDM women were divided into an intervention and a control group (28 women in each group), with the former receiving omega-3 fatty acid supplementation for 6 weeks. Again, neither the intervention nor the control group showed amelioration in the HOMA beta-cell index. However, some results may be influenced by the short duration of intervention and the low dose of omega-3 fats. In fact, despite the known effect of omega-3 fats in lowering serum triglycerides, in this study such effect was not found either.

The study by Horie et al. [14] was a descriptive analysis of the effects of medical nutrition therapy on GDM. All women had early diagnosis of GDM (prior to 20 weeks). Women were instructed to take six meals per day, in which the carbohydrate content was divided into three small main meals and three snacks. Women underwent a 75-g OGTT at early gestation period, at mid-gestation (at least 4 weeks after nutritional therapy), and 4–8 weeks postpartum. By mid-gestation, 18 of the 41 women treated with medical nutrition therapy normalized their glucose tolerance. As regards the remaining women, 9 of them required insulin during the pregnancy; also, 9 of them developed impaired glucose tolerance or diabetes after delivery. Beta-cell function was evaluated with three indices: the 1 hour to fasting insulin difference over the same glucose difference, the insulin to glucose AUC ratio, and the oral disposition index (though computed slightly differently than in study [12]). Results were presented with women stratified into the two groups indicated above, i.e., women who at mid-pregnancy reverted to normal glucose tolerance, versus those women whose GDM condition remained. Only in the former, some of the indices showed an improvement in beta-cell function due to nutrition therapy, though results were not completely consistent.

In one study, women received both nutrition and, possibly, pharmacological therapy.[15] It was a randomized study, where 275 women underwent intensive or standard therapy (127 and 148 women, respectively). The intensive therapy group received individualized dietary and exercise advice, whereas the standard group received general education only. Women in both groups could be treated with insulin, but only in the case of fasting or 2-hour post-prandial glucose above a prescribed threshold. However, the intensive group was asked to perform regular self-monitoring of glucose, whereas in the standard group, self-monitoring was advised but not required. Women were also revaluated 1–3 years after delivery, but only approximately 40 women in each group were examined. In that occasion, an OGTT was performed, and beta-cell function was evaluated by the insulinogenic index (30 min minus fasting insulin, normalized to the same difference in glucose), and by the HOMA beta-cell index. No significant difference in beta-cell function was found between the women that had previously received intensive or standard therapy.

4. Conclusion

The current treatment for GDM includes medical nutrition therapy, lifestyle modification and possible use of pharmacological agents or nutrition supplementation. One aspect in the assessment of the metabolic condition of women with GDM, which has gained increasing relevance in the last years, is pancreatic beta-cell function. However, very few studies reported quantitative information about the effects on beta-cell function of the treatment of GDM, despite the fact that some agents, such as glyburide, are well known insulin secretagogues. Of note, to our knowledge even less information is available on fetal beta-cell function, in GDM women under pharmacological treatment.

5. Expert opinion

The effect of treatments for GDM on insulin sensitivity is well known. At contrast, few studies provided information on the effects on beta-cell function. A possible reason is that the role of beta-cell function deterioration in GDM has been clearly demonstrated only in recent years. Of note, even the studies that reported some quantitative data on beta-cell function effects of treatment, which are summarized in Table 1, typically presented basic information, obtained by simple, empirical indices of beta-cell function, such as the insulinogenic index, or the HOMA beta-cell index. Only the study by Hebert et al. [8] presented more accurate results on beta-cell function, based on mathematical modeling. However, even this study [8] had limitations, due to the relatively low number of the GDM women, and to the type of test performed, which was a mixed meal tolerance test instead of an OGTT. In fact, OGTT is the standard procedure for diagnosing GDM, though it may be ethically questionable to repeat this examination in already diagnosed GDM women. However, in our opinion, reliable assessment of beta-cell function would require the OGTT rather than the mixed meal test, the latter not being completely standardized yet. In summary, further studies should be carried out for an improved knowledge of the effects of treatments on beta-cell function, which, together with the assessment of insulin sensitivity, would likely result in more appropriate, thus optimized, therapeutic approaches for the management of women with GDM, both during pregnancy and possibly after delivery.

Table 1. Main characteristics of studies with quantitative assessment of the effects of treatment on pancreatic beta-cell function in GDM.

Authors and ref. number	Number of gestational diabetic women	Type of treatment	Type of beta-cell function assessment
Hebert et al. [8]	40 women	Glyburide	Model-based indices (beta-cell responsivity, disposition index), based on mixed meal test
Glueck et al. [9]	3 of 39 polycystic ovary syndrome women	Metformin	Empirical index (HOMA beta-cell)
Xiang et al. [10]	27 women, studied postpartum	Pioglitazone	Empirical index (acute insulin response and disposition index, based on intravenous glucose tolerance test)
Mozaffari-Khosravi et al. [11]	24 women	Vitamin D	Empirical index (HOMA beta-cell)
Yeow et al. TP [12]	26 women, studied postpartum	Vitamin D	Empirical indices (disposition, insulinogenic-like, and acute insulin response indices, based on OGTT)
Samimi et al. [13]	28 women	Omega-3 fatty acid	Empirical index (HOMA beta-cell)
Horie et al. [14]	41 women	Nutrition therapy	Empirical indices (disposition, and insulinogenic-like indices, based on OGTT)
Cao et al. [15]	81 women	Nutrition therapy and possibly insulin	Empirical indices (insulinogenic index, based on OGTT, and HOMA beta-cell)

Declaration of Interest

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending or royalties.