The reported frequency of use in pregnancy and during breastfeeding for most of these agents is very low or completely absent.
The five subclasses of urinary tract agents are analgesics, antispasmodics, urinary acidifiers, urinary alkalinizers, and urinary germicides. With the exception of the three urinary germicides, anti-infectives are not covered in this column.
The analgesic subclass includes pentosan and phenazopyridine. Pentosan (Elmiron), a heparinlike compound, is an oral drug that is indicated for the relief of bladder pain or discomfort associated with interstitial cystitis. Systemic absorption is low, at about 6%. Because of the high molecular weight (4,000-6,000), it does not appear to cross the placenta, at least in the first half of pregnancy. A 1975 reference described its use in five women with preeclampsia. Each patient received 100 mg intramuscularly every 8 hours for about 5 days in the last weeks of pregnancy. No maternal benefit from the therapy was observed. There was apparently no fetal harm, but the neonatal outcomes were not described.
There are substantial – more than 900 – human pregnancy exposures in the first trimester with phenazopyridine. The exposures were not related to an increased risk of embryo-fetal harm and so use of the drug in pregnancy can be classified as compatible. However, the low molecular weight (about 214 for the free base) suggests that the drug will cross to the embryo and fetus.
The eight antispasmodics are darifenacin (Enablex), fesoterodine (Toviaz), flavoxate, mirabegron (Myrbetriq), oxybutynin (Ditropan XL), solifenacin (Vesicare), tolterodine (Detrol LA), and trospium.
These agents are indicated for the treatment of overactive bladder with symptoms of urge urinary incontinence, urgency, and frequency. The molecular weights range between 342 and 508, suggesting that all will cross the human placenta. There are no human pregnancy data for six of these agents and very limited data for flavoxate and oxybutynin. There is no evidence of embryo-fetal harm from these two drugs, but only one case involved exposure in the first trimester.
In seven of these drugs, the animal data suggested low risk. There was no embryo harm from doses that were equal to or less than 10 times the human dose based on body surface area (BSA) or area under the concentration curve (AUC). Solifenacin did cause embryo toxicity in pregnant mice. There was no embryo toxicity in pregnant rats and rabbits, but the maximum doses used were very low. Overall, the available data suggest that exposure to an antispasmodic in pregnancy is low risk for embryo, fetal, and newborn harm.
Ammonium chloride is a urinary acidifier as well as a respiratory expectorant. There is a large amount of data related to when the drug was used as an expectorant. There was no evidence that this use was associated with large categories of major or minor malformations. However, there were possible associations with three individual defects: inguinal hernia, cataract, and any benign tumor. No reports describing its use as a urinary acidifier have been located. When large amounts are consumed near term, the drug may cause acidosis in the mother and fetus. The molecular weight (about 53) suggests that it will cross the placenta.
Potassium citrate (Urocit-K) is indicated for the management of renal tubular acidosis with calcium stones, hypocitraturic calcium oxalate with nephrolithiasis of any etiology, and uric acid lithiasis with or without calcium stones. The molecular weight (about 307) suggests it will cross the placenta. Only one case of its use in pregnancy has been located. The newborn had an unspecified defect but no other information was provided. The animal data in four species suggest low risk.
There are three urinary germicides: methenamine, methylene blue, and nitrofurantoin. Methenamine is available as methenamine mandelate (molecular weight abut 292) and methenamine hippurate (molecular weight about 319). Both are metabolized to formaldehyde (molecular weight about 30), the active agent. The molecular weights suggest that all will cross the placenta. The use of methenamine during pregnancy has been reported in more than 750 pregnancies. There have been no embryo or fetal adverse effects attributed to the drug.
The human data involving oral methylene blue, a weak urinary germicide, is limited to 55 exposures. There were three infants with birth defects (type not specified). Several reports have described the use of intra-amniotic injections to assist in the diagnosis of suspected membrane rupture. This use has resulted in newborns with hemolytic anemia, hyperbilirubinemia with or without Heinz body formation, blue staining of the skin, and methemoglobinemia. Fetal deaths have also been described. Recommendations to avoid the intra-amniotic use of methylene blue were issued more than 10 years ago. Moreover, the use of oral methylene blue as a urinary germicide is no longer recommended.
The low molecular weight (about 238) of nitrofurantoin suggests that it will cross the placenta. It is commonly used in pregnancy for the treatment or prophylaxis of urinary tract infections. The large amount of human data indicates that the risk of drug-induced birth defects is low. Several cohort studies have found no increased risk for birth defects. However, some case-control studies have found increased risks for hypoplastic left heart syndrome and oral clefts. A 2015 review concluded that this difference was due to the increased sensitivity of case-control studies to detect adverse effects ().
Use of the drug close to term may cause hemolytic anemia in newborns who are glucose-6-phosphate dehydrogenase (G6PD) deficient. Although rare, this may also occur in newborns who are not G6PD deficient. The best course is to avoid use of the drug close to delivery. As for use of the drug in the first trimester, ACOG’s Committee on Obstetric Practice stated in Committee Opinion No. 717 that nitrofurantoin was still thought to be appropriate when no other suitable alternative antibiotics were available (
Except for methenamine and nitrofurantoin, there are no data related to the use of the urinary tract drugs during breastfeeding. Peak levels of methenamine occur at 1 hour, but no reports of adverse effects on nursing infants have been located. Several reports have described the use of nitrofurantoin during breastfeeding. Minor diarrhea was noted in two infants. However, breastfeeding an infant with G6PD deficiency could lead to hemolytic anemia.
Phenazopyridine should be used with caution especially for an infant younger than 1 month or with G6PD deficiency because of the risk for methemoglobinemia, sulfhemoglobinemia, and hemolytic anemia.
Although there have been no reports of the use of mirabegron during lactation, the characteristics of the drug – low molecular weight (about 397), long elimination half life (50 hours), and moderate plasma protein binding (about 71%) – suggest that the drug will be excreted into milk, potentially in clinically significant amounts. There is also concern with use of tolterodine (molecular weight about 476) because both the primary drug and its equipotent metabolite may be excreted into milk.
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.