ORIGINAL ARTICLE Pharmacological Management of Gestational Diabetes An overview DONALD R. COUSTAN, MD presumably the fetus is relatively pro- tected from any direct effects. An example of such a medication would be heparin, OBJECTIVE — To provide a review of the background literature regarding the pharmaco- which as a relatively large and highly logical management of gestational diabetes. charged molecule, does not traverse the placenta to any measurable extent. Hepa- RESEARCH DESIGN AND METHODS — This is a literature review. rin is generally considered to be safe for the fetus, although of course it poses sig- RESULTS — Information is available regarding the use of some, but not all, oral antidiabetes nificant risks such as hemorrhage and agents in pregnancy. bone demineralization for the mother. CONCLUSIONS — Available evidence supports the use of glyburide during pregnancy. If a medication is shown to cross the Evidence is inadequate to support or refute the use of metformin, an agent that has been shown placenta, that is not necessarily a contra- to cross the placenta and thus could be helpful or harmful to the developing fetus. indication to its use in pregnancy. For example, we frequently prescribe dexa- Diabetes Care 30 (Suppl. 2):S206–S208, 2007 methasone or betamethasone to pregnant women to take advantage of transplacen- ecause insulin preparations tested to GUIDING PRINCIPLES — As our tal passage of these drugs, to enhance fetal date have been determined not to understanding of the physiology and lung maturation. The second question is B cross the placenta or to cross mini- pharmacology of both pregnancy and di- whether a medication that reaches the fe- mally, insulin has been the treatment of abetes has increased, some guiding prin- tal compartment is neutral, helpful, or choice in most parts of the world for pa- ciples have emerged to help us decide harmful to the fetus. tients with gestational diabetes whose which medications are safe and effective The second safety question above also circulating glucose levels exceed prede- for pregnant women. Any medication touches on the subject of efficacy. If a termined thresholds. Although advances must, of course, be safe for the person drug crosses the placenta and is helpful to have been made in developing insulins who takes it, or at least the benefits of its the fetus without causing harm, it could that may be administered by alternative use must outweigh the risks. However, then be considered to be both safe and routes, insulin is typically injected subcu- when a pregnant woman takes a medica- efficacious. Most often, a drug is pre- taneously. This poses a barrier to utiliza- tion, we are also concerned about that scribed during pregnancy to help the tion and has been one of the factors that medication’s effects on her unborn fetus. mother. The first efficacy question is kindled interest over many years in treat- Starting with the thalidomide experience whether the drug has been shown to help ing gestational diabetes with oral antidia- in the 1960s, our view of the placenta has the condition for which the mother is be- betic agents. This and the following shifted from that of a barrier, serving to ing treated. If the mother has diabetes, presentations describe oral agents that protect the fetus from harm, to that of a and we are considering oral antidiabetic have the potential to be used as alterna- sieve, allowing entry of all kinds of chem- drugs, the answer would generally be yes. tives to insulin for the treatment of gesta- icals and substances to the fetal circula- However, this will not always be the case. tional diabetes. In this overview, I shall tion. Whereas neither point of view is A medication whose mechanism of action outline some guiding principles in deter- entirely accurate, the latter is probably a is to stimulate pancreatic insulin secretion mining whether a particular agent is ap- safer perspective. Thus, the first question and/or release would not be helpful to a propriate for use in pregnancy and then one should ask about a medication to be mother with type 1 diabetes. apply these principles to the most com- used by pregnant women is whether it There are situations where treating monly used oral antidiabetes agents. crosses the placenta. If it does not, then the mother’s condition will also help the ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● fetus indirectly. The second efficacy ques- From the Department of Obstetrics and Gynecology, Brown Medical School, and the Women & Infants tion is whether the drug’s effect on the Hospital of Rhode Island, Providence, Rhode Island. mother will be beneficial, and not harm- Address correspondence and reprint requests to Donald R. Coustan, Chace/Joukowsky Professor and ful, to the fetus. In keeping with the Ped- Chair, Department of Obstetrics and Gynecology, Brown Medical School, Obstetrician and Gynecologist in ersen Hypothesis, any medication that Chief, Women & Infants Hospital of Rhode Island, 101 Dudley St., Providence, RI 02905-2401 E-mail: [email protected]. tends to normalize maternal glycemic lev- Received for publication 28 March 2006 and accepted in revised form 14 June 2006. els should benefit the fetus as well. The This article is based on a presentation at a symposium. The symposium and the publication of this article third efficacy question is whether the were made possible by an unrestricted educational grant from LifeScan, Inc., a Johnson & Johnson company. drug crosses the placenta and has a direct A table elsewhere in this issue shows conventional and Syste`me International (SI) units and conversion factors for many substances. benefit to the fetus. DOI: 10.2337/dc07-s217 It is my view that all of these questions © 2007 by the American Diabetes Association. should be considered in making a deci- S206 DIABETES CARE, VOLUME 30, SUPPLEMENT 2, JULY 2007 Coustan sion regarding treatment of gestational di- detected in the cord blood of offspring latter statement. However, Hague et al. abetes with oral antidiabetes agents. whose mothers took the drug as part of a (8) measured plasma metformin levels in randomized trial (4). No information is seven women taking metformin at a me- CLASSES OF ORAL available regarding nateglinide; the pack- dian daily dose of 2,000 mg and in the ANTIDIABETES AGENTS — Oral age insert (5) lists the drug as Pregnancy cord blood of 23 babies whose mothers antidiabetes agents are typically classified Category C, but states, “There are no ad- took metformin during pregnancy. Me- as insulin secretagogues, insulin sensitiz- equate and well-controlled studies in dian plasma metformin levels were 1.05 ers, and ␣-glucosidase inhibitors. Re- pregnant women. Starlix should not be ␮g/ml (range 0.06–2.93) in maternal cently, a glucagon-like peptide 1 agonist used in pregnancy.” For repaglinide, the blood and 0.63 ␮g/ml (range 0.08–2.55) has also been placed on the market. Physicians’ Desk Reference (6) states, in cord blood samples. These data suggest “Prandin should be used during preg- that significant amounts of metformin can Insulin secretagogues nancy only if it is clearly needed.” Given cross the placenta, with fetal concentra- The sulfonylureas and meglitinide are in- the available data, glyburide appears to be tions in the range of half of maternal con- sulin secretagogues. The sulfonylureas the best candidate insulin secretagogue centrations. Because it is unknown bind to sulfonylurea receptors in ␤-cells, for use during pregnancy, since it crosses whether metformin is therapeutic or del- stimulating insulin secretion at all blood the placental little or not at all and bene- eterious to the fetus, it would seem pru- glucose levels. For sulfonylureas to be ef- fits the mother directly and the fetus indi- dent to obtain further data (perhaps from fective, the patient must have residual rectly. A subsequent presentation will go animal models) before metformin be- ␤-cell function, so these drugs are not at into detail regarding the use of this drug comes commonly prescribed during all effective in patients with type 1 diabe- for women with gestational diabetes. pregnancy. At the very least, patients tak- tes or longstanding type 2 diabetes in the ing metformin should be counseled about stage of insulinopenia. The primary side Insulin sensitizers the unknown risks and benefits for the effect of sulfonylureas is hypoglycemia. There are two broad types of insulin sen- fetus. The effect of sulfonylureas is to suppress sitizers: the biguanides and the thiazo- The thiazolidinediones are agonists hepatic glucose production, diminish lidinediones, also called peroxisome for the peroxisome proliferator–activated glucotoxicity, and improve insulin secre- proliferator–activated receptor-␥ agonists. receptor-␥. Such receptors are found in tion after meals. They generally lower cir- The biguanides enhance insulin ac- target tissues for insulin action. These culating glucose levels by ϳ20% and tion, stimulating glucose uptake in the drugs enhance peripheral insulin action work best in patients of normal or slightly liver and in the periphery and also sup- and are useful for patients with type 2 increased body weight. The first- pressing hepatic glucose output. They diabetes who have adequate endogenous generation sulfonylureas include tolbut- only work when insulin is present, do not insulin; they are only useful if insulin re- amide, chlorpropamide, and tolazamide. stimulate insulin secretion or release, and sistance is present. Weight gain is com- Second-generation sulfonylureas include do not cause hypoglycemia. They are mon with these drugs and appears to be glipizide (Glucotrol), which is shorter act- used for patients with type 2 diabetes who dose and time dependent. Fluid retention ing; glyburide or glibenclamide (Diabeta have residual ␤-cell function, typically may occur, and peripheral edema devel- and Micronase), which are longer acting; when diet and exercise are insufficient for ops in 2–5% of patients. Heart failure may and gimepride (Amaryl), which is also diabetic control.