Drugs in Pregnancy and Lactation: Tenth Edition


Muscle Relaxant

PREGNANCY RECOMMENDATION: Limited Human Data—No Relevant Animal Data

BREASTFEEDING RECOMMENDATION: Limited Human Data—Probably Compatible


The limited human pregnancy experience with carisoprodol does not suggest a major risk of embryo or fetal harm. The FDA data do raise a hypothesis of an association with oral clefts, but the risk, if it exists, appears to be low. There have been some reports of congenital defects with meprobamate, the active metabolite of carisoprodol, but two large surveillance studies did not support an association between meprobamate and malformations (see Meprobamate). Although the data do not support a significant risk with carisoprodol, the best course would be to avoid this agent, if possible, during the 1st trimester.


Carisoprodol is a centrally acting muscle relaxant. It is metabolized by the liver to meprobamate, the active metabolite. The reproductive effects of this drug in animals have not been studied.

It is not known if carisoprodol crosses the human placenta. The molecular weight (about 260) suggests that it will cross to the embryo and fetus. Moreover, the active metabolite, meprobamate, does cross the placenta (see Meprobamate).

The Collaborative Perinatal Project monitored 50,282 mother–child pairs, 14 of whom were exposed in the 1st trimester to carisoprodol (1). No association of the drug with large classes of malformations or to individual defects was found.

In a surveillance study of Michigan Medicaid recipients involving 229,101 completed pregnancies conducted between 1985 and 1992, 326 newborns had been exposed to carisoprodol during the 1st trimester (F. Rosa, personal communication, FDA, 1993). Twenty (6.1%) major birth defects were observed (14 expected), including (observed/expected) 3/3 cardiovascular defects, 2/0.5 oral clefts, and 1/1 hypospadias. No anomalies were observed in three other categories of defects (spina bifida, polydactyly, and limb-reduction defects) for which data were available. Only with the two cases of oral clefts is there a suggestion of a possible association, but other factors, including the mother’s disease, concurrent drug use, and chance, may be involved.

A 2001 case report described the pregnancy outcome of a 27-year-old woman who had taken carisoprodol (700 mg), propoxyphene (70 mg), and acetaminophen (900 mg) three times daily throughout gestation and during the first 6 months of breastfeeding (2). The 2635-g infant (sex not specified), delivered 2 weeks preterm by cesarean section, showed no signs or symptoms of withdrawal after birth. No mention was made of birth defects, and the infant was developing normally at 6 months of age (2).

A similar case was observed in 2006 (3). The woman took carisoprodol (700 mg) four times daily before and throughout gestation for severe back muscle spasm. A healthy female infant was delivered at term. The mother continued treatment while breastfeeding during the first postpartum month. No signs or symptoms of withdrawal were noted in the newborn or in the infant when breastfeeding was stopped about 1 month after birth (3).


Carisoprodol and its active metabolite, meprobamate, are excreted into breast milk at concentrations two to four times those in the maternal plasma (4). While unique, the two cases below confirm that the drug and its metabolite are present in milk and that the concentrations of meprobamate exceed those in the maternal plasma.

A 2001 case report described the use of carisoprodol (700 mg), propoxyphene (70 mg), and acetaminophen (900 mg) three times daily throughout gestation and during the first 6 months of breastfeeding (2). Levels of carisoprodol and meprobamate were measured in breast milk on 4 consecutive days. The average milk concentrations of the two agents were 0.9 mcg/mL and 11.6 mcg/mL, respectively. Neither the timing of the samples in relationship to the maternal dose nor the maternal plasma levels were specified. Based on an estimated milk intake of 150 mL/kg/day, the absolute and relative doses of carisoprodol plus meprobamate that would have been ingested by an exclusively breastfed infant were 1.9 mg/kg/day and 4.1%, respectively. No signs or symptoms of withdrawal were noted in the infant who had normal psychomotor development at 6 months of age. However, the infant was partially fed formula because of insufficient maternal milk production (2).

In a similar case, a woman took carisoprodol (700 mg) four times daily before and throughout gestation, and during the first postpartum month for severe back muscle spasm (3). The healthy female infant was delivered at term and was exclusively breastfed. The mother also took an analgesic combination (hydrocodone plus acetaminophen) for a short period after delivery. Blood and milk samples were obtained from the mother about a week after delivery. Peak carisoprodol and meprobamate blood concentrations, obtained 2 hours post-dose, were 3 and 9 mcg/mL, respectively, whereas the milk concentrations, obtained at the same time, were 1.4 and 10.9 mcg/mL, respectively. The corresponding milk plasma ratios were 0.3 and 1.4, respectively. Trough blood concentrations, obtained about 4 hours postdose, were <2 and <4 mcg/mL, respectively, whereas the milk values were 0.8 and 17.1 mcg/mL, respectively. The corresponding milk:plasma ratios were ≥0.4 and ≥4.3, respectively. A blood sample from the infant, obtained 2 hours after breastfeeding that was started 1.5 hours after a maternal dose, revealed undetectable levels of carisoprodol (<2 mcg/mL) and meprobamate (<4 mcg/mL). Nursing was stopped at about 1 month of age. No signs or symptoms of withdrawal or other adverse effects were noted in the newborn or when breastfeeding was stopped about 1 month later (3).

Although the human data are very limited, both cases involved mothers taking carisoprodol doses that were much higher than the recommended dose of 1050–1400 mg/day. The lack of detectable adverse effects in both infants suggests that the risk of toxicity is low, at least in infants that also were exposed during pregnancy. Starting carisoprodol treatment during breastfeeding may have different results in a nursing infant. The American Academy of Pediatrics classifies another centrally acting skeletal muscle relaxant as compatible with breastfeeding (see Baclofen). However, until additional data are available, women taking this drug and who elect to nurse should closely monitor their infants for sedation and other changes in behavior or functions.


1.Heinonen OP, Slone D, Shapiro S. Birth Defects and Drugs in Pregnancy. Littleton, MA: Publishing Sciences Group, 1977:357–65.

2.Nordeng H, Zahlsen K, Spigset O. Transfer of carisoprodol to breast milk. Ther Drug Monit 2001;23:298–300.

3.Briggs GG, Ambrose PJ, Nageotte MP, Padilla G. High-dose carisoprodol during pregnancy and lactation. Ann Pharmacother 2008:42:898–901.

4.Product information. Soma. Wallace Laboratories, 1994.