The three most potent human teratogens, with the possible inclusion of some of the first antineoplastics, are isotretinoin, alcohol, and hyperglycemia.
As with all teratogens, the toxicity is dose related. For example, the risk of embryo-fetal harm from hyperglycemia increases markedly when the HbA1c is greater than 8%. Moreover, diabetes accounts for more than 90% of the harm caused by chronic diseases. Consequently, control of glucose levels in pregnancy is critical.
If these agents are used near term, there is a risk that they will cause hypoglycemia in the newborn. Changing from oral therapy to insulin is the safest course.
There are seven pharmacologic subclasses of oral antidiabetic agents: alpha-glucosidase inhibitors, biguanides, dipeptidyl peptidase-4 inhibitors, meglitinides, sulfonylureas, sodium-glucose cotransporter-2 inhibitors, and thiazolidinediones. Many of these drugs are available in combination with metformin. All of these agents are indicated as adjunct to diet and exercise for type 2 diabetes, but they also can be used for gestational diabetes. Although the human pregnancy data are very limited or nonexistent for most of these agents, none are known to cause structural defects in humans. Additional details of the exposures are available in the 11th edition of “” (2017: Wolters Kluwer).
The two agents is this subclass are acarbose (Precose) and miglitol (Glyset). The human pregnancy data with acarbose are limited, and no human pregnancy data have been found for miglitol. The animal data for both drugs suggest low risk.
There are substantial human pregnancy data for metformin in both type 2 and gestational diabetes. When combined with insulin, it is effective in significantly lowering the amount of insulin required to control hyperglycemia. It also may be effective when used alone. The risk of embryo-fetal harm with this drug appears to be very low or nonexistent. The animal data suggest low risk.
Dipeptidyl peptidase-4 inhibitors
There are four drugs in this subclass: alogliptin (Nesina), linagliptin (Tradjenta), saxagliptin (Onglyza), and sitagliptin (Januvia). No reports of the use of the first three drugs in human pregnancy have been found. However, the(2006-2009) described the outcomes of eight women who were exposed to sitagliptin or sitagliptin/metformin in the first trimester. The outcomes of these pregnancies were five healthy newborns, two spontaneous abortions, and one fetal death at 34 weeks’ gestation. In that case, the mother took sitagliptin and metformin separately during the first 5 weeks of gestation. The animal data for all four drugs suggest low risk.
Nateglinide (Starlix) and repaglinide (Prandin) are the agents in this subclass. There is no human pregnancy data for nateglinide, but there is limited data (eight pregnancies) for repaglinide. No birth defects or other toxicity was noted in these cases. The animal data suggest low risk.
Six drugs are included in this subclass: chlorpropamide, glimepiride (Amaryl), glipizide (Glucotrol), glyburide, tolazamide (Tolinase), and tolbutamide. These agents were among the first oral antidiabetic agents. As a result, they have the most human pregnancy data. Although birth defects were observed in newborns of mothers who had used one of these drugs, the defects were thought to be the result of uncontrolled diabetes. The animal data suggest low risk.
There are three drugs in this sodium-glucose cotransporter-2 inhibitor subclass: canagliflozin (Invokana), dapagliflozin (Farxiga), and empagliflozin (Jardiance). No reports describing the use of these drugs in human pregnancy have been located. The animal data suggest low risk.
Pioglitazone (Actos) and rosiglitazone (Avandia) form this subclass. There are limited human pregnancy data for both drugs. The animal data suggest moderate risk for embryo-fetal toxicity but not for structural defects.
All of the above drugs will probably be excreted into breast milk, but the amounts are typically unknown. When they have been measured, the amounts were usually low. However, there is still a risk for hypoglycemia in a nursing infant. Combination products containing two antidiabetic agents are best avoided. The safest course is to use insulin, but, if this is not an option, then the lowest effective dose should be used. In addition, the infant’s blood glucose levels should be routinely monitored.
Mr. Briggs is a clinical professor of pharmacy at the University of California, San Francisco, and an adjunct professor of pharmacy at the University of Southern California, Los Angeles, as well as at Washington State University, Spokane. He coauthored “Drugs in Pregnancy and Lactation” and coedited “Diseases, Complications, and Drug Therapy in Obstetrics.” He reported having no relevant financial disclosures.