Drugs in Pregnancy and Lactation: Tenth Edition



PREGNANCY RECOMMENDATION: Limited Human Data—Animal Data Suggest Risk



The animal data suggest risk, but the limited human pregnancy experience with levetiracetam has not shown an increased risk of major congenital malformations. Although confirmation is needed, severe growth restriction has been observed. Additional data are required before a more complete assessment of embryo–fetal risk can be derived. It is not known if levetiracetam causes folic acid deficiency, but it will usually be combined with other anticonvulsants, some of which may cause folic acid deficiency. Therefore, daily supplementation with 4–5 mg of folic acid, combined with multivitamins that include adequate amounts of other B vitamins, is recommended. If a pregnant patient is receiving levetiracetam, clinicians are encouraged to register her, before pregnancy outcome is known, in the Antiepileptic Drug Pregnancy Registry by calling the toll-free number (888-233-2334) (1).


Levetiracetam is an anticonvulsant used as adjunctive therapy in the treatment of partial-onset seizures in epilepsy. It is chemically unrelated to other anticonvulsant agents. Levetiracetam is minimally protein bound (<10%) and is not metabolized by the liver (1,2). Major metabolism occurs by enzymatic hydrolysis of the acetamide group to produce the carboxylic acid metabolite (1). Levetiracetam does not inhibit and is not a high affinity substrate for epoxide hydrolase (1). Reports describing the effects (if any) of levetiracetam on folic acid have not been located.

In reproduction studies with pregnant rats, maternal doses approximately equivalent to the maximum recommended human dose (3000 mg) based on BSA (MRHD) administered throughout pregnancy and lactation were associated with an increased incidence of minor fetal skeletal abnormalities and restricted growth prenatally and postnatally (1). When doses six times the MRHD were used, increased pup mortality and behavioral alterations were observed. The developmental no-effect dose administered throughout pregnancy and lactation was 0.2 times the MRHD. Dosing at 12 times the MRHD given only during organogenesis resulted in reduced fetal weights and an increased incidence of fetal skeletal variations. The developmental no-effect dose during organogenesis was four times the MRHD. No evidence of fetal or neonatal harm was observed with doses up to six times the MRHD during the last third of pregnancy and throughout lactation. None of the studied doses produced maternal toxicity (1).

In pregnant rabbits, increased embryo/fetal mortality and an increased incidence of minor skeletal abnormalities were observed at maternal doses four times the MRHD administered during organogenesis. When a dose 12 times the MRHD was used, decreased fetal weight and increased incidences of fetal malformations were noted, but maternal toxicity was also evident at this dose. The developmental no-effect dose was 1.3 times the MRHD (1).

It is not known if levetiracetam crosses the human placenta. The low molecular weight (about 170) and the lack of protein binding suggest that exposure of the embryo and fetus should be expected.

An estimated 20%–30% of epileptic patients have seizures not well controlled by monotherapy and require anticonvulsant polytherapy (2). Clinical studies of levetiracetam have demonstrated efficacy in partial-onset seizures with a good tolerability profile. In addition, its lack of hepatic metabolism and low protein binding result in a low risk of interaction with other drugs, including other anticonvulsants and oral contraceptives. Although comparison studies with other adjunctive anticonvulsants (e.g., felbamate, gabapentin, lamotrigine, tiagabine, topiramate, and vigabitrin) have not been conducted, it should be added to the list of agents to consider in cases of treatment-refractory partial-onset seizures (2).

A 2005 report described the outcomes of 11 pregnancies exposed to levetiracetam that were enrolled in the European Registry of Antiepileptic Drugs and Pregnancy (3). Levetiracetam was used alone in two pregnancies and combined with other antiepileptic agents in nine. The outcomes were one spontaneous abortion, one elective abortion, and nine infants without birth defects. However, three of the infants, one of whom was premature, had very low birth weight (≤5th percentile) (3).

The Lamotrigine Pregnancy Registry, an ongoing project conducted by the manufacturer, was first published in January 1997 (4). The final report was published in July 2010. The Registry is now closed. Among 57 prospectively enrolled pregnancies exposed to levetiracetam and lamotrigine, with or without other anticonvulsants, 54 were exposed in the 1st trimester resulting in 48 live births without birth defects, 3 SABs, 1 fetal death, and 2 birth defects. There were three exposures in the 2nd/3rd trimesters resulting in live births without defects (4).


No reports describing the use of levetiracetam during lactation have been located. The molecular weight (about 170) and low protein binding (<10%) suggest that levetiracetam will be excreted into breast milk. The effects of this exposure on a nursing infant are unknown. Of note, however, other anticonvulsants (see Carbamazepine, Phenytoin, and Valproic Acid) are classified as compatible with breastfeeding by the American Academy of Pediatrics.


1.Product information. Keppra. UCB Pharma, 2002.

2.Dooley M, Plosker GL. Levetiracetam. A review of its adjunctive use in the management of partial-onset seizures. Drugs 2000;60:871–93.

3.Ten Berg K, Samren EB, van Oppen AC, Engelsman M, Lindhout D. Levetiracetam use and pregnancy outcome. Reprod Toxicol 2005;20:175–8.

4.The Lamotrigine Pregnancy Registry. Final Report. 1 September 1992 through 31 March 2010. GlaxoSmithKline, July 2010.