This site is intended for health professionals only

Optimal management of epilepsy during pregnancy

Marianna V Spanaki
Assistant Professor of Neurology

Santiago Arroyo
Associate Professor of Neurology and Director of the Comprehensive Epilepsy Program
Neurology Department
Medical College of Wisconsin

Epilepsy is a common neurological problem with an incidence rate of approximately 0.5%. Women with epilepsy represent a distinct patient population with unique healthcare needs. Coordinating epilepsy management becomes even more critical in women with epilepsy who are of childbearing age and eventually become pregnant. More than 90% of women with epilepsy have normal healthy pregnancy outcomes.(1)

However, pregnancy in women with epilepsy who are taking antiepileptic drugs (AEDs) should be considered as high risk due to:

  • Increased risk of seizures in about one-third of women.(2)
  • Variations in the serum concentrations of AEDs secondary to ­pregnancy-related ­pharmacokinetic changes.(3)
  • Increased likelihood of obstetric complications.(4)
  • Increased risk of birth defects.(5,6)

The characteristics of AEDs during pregnancy are shown in Table 1.


Pregnancy in women with epilepsy should always be planned, and ideally proper management should be instituted before conception. It is important to re-investigate the epileptic syndrome thoroughly before conception using the available techniques (electroencephalogram and brain magnetic resonance imaging [MRI]). Lesions that may complicate pregnancy, such as aneurysm, tumours, cavernous haemangioma or arterio­venous malformations, should be ruled out.

Increased seizures can occur in about a third of pregnant epileptic women.(2) Sleep deprivation and noncompliance are major causes of breakthrough seizures. In addition, the pharmacokinetic parameters of AEDs are altered in pregnancy due to:(3)

  • Increased plasma volume and cardiac output.
  • Increased hepatic metabolism.
  • Increased glomerular filtration rate and renal clearance.
  • Decreased protein binding due to decreased albumin.

It has been shown that, due to changes in plasma protein binding, there is a significant reduction in the total concentration of all the classic AEDs (phenytoin, carbamazepine, valproate and phenobarbital) during pregnancy.(3) However, unbound levels decline significantly only for phenobarbitol. Levels of free valproate increase by 25%. The changes are more pronounced towards the end of the third trimester and the immediate postpartum period. Little is known about the pharmacokinetic changes of the novel AEDs since very few exposures have been reported. Lamotrigine levels steadily decline during pregnancy and return to ­baseline levels two to three weeks postpartum.(7)

The increased risk of seizures due to changes in AED concentration can be prevented by closely monitoring AED free levels and adjusting the dosage. As a rule, AED levels should be monitored before conception, at the beginning of each trimester and during the last month of pregnancy.

Use of AEDs during conception and pregnancy has been associated with increased risk of congenital malformations.(1,5,6) In women who are medication-free the estimated rate of birth defects is 2–3%. The incidence of malformations in infants of mothers with epilepsy is likely to be two to three times that of infants of mothers not taking medication (4–6%). Congenital defects are most likely caused by exposure of infants to AEDs in utero and not by the maternal underlying epilepsy.(6) Specific risk factors include elevated maternal daily dosage or serum concentration of AEDs, polytherapy and low folate levels. Malformations ­identified in infants exposed to AEDs include:

  • Malformations of the mid face and digits in 5–30%.
  • Cleft lip, palate and cardiac ventricular septal defects in 4–8%.
  • Neural tube defects in 1–2%. There is significant evidence that valproate is associated with 1–2% and carbamazepine with 0.5% risk of spina bifida. The risk for valproate is diminished if the total daily dosage does not exceed 1,000mg and is given in four divided doses.

Little information is available regarding the safety of the newer AEDs (lamotrigine, levetiracetam, gabapentin, tiagabine, oxcarbazepine, zonisamide, felbamate and topiramate).

Lamotrigine inhibits the release of excitatory amino acids and has mild anti­folate activity. There have been three outcomes with major defects on lamotrigine monotherapy among 120 outcomes (2.5%) involving a first-trimester exposure. There were 13 outcomes with major defects among 275 total outcomes (10.4%) involving first-trimester exposure to lamotrigine combined with other AEDs.

Levetiracetam is not metabolised by hepatic cytochrome P450 and yields an inactive primary metabolite. Of 28 pregnancies reported on levetiracetam either as mono­therapy or polytherapy, 13 were incomplete, 13 of the remaining 15 had favourable outcomes, three were associated with congenital malformations and one had an unknown outcome.

Gabapentin, a gamma-aminobutyric acid (GABA)-mimetic drug, has no reactive metabolites, which may indicate that it is safer during pregnancy. Of ten known pregnancy outcomes, four were terminated by elective abortion and six healthy infants were born.

Tiagabine, a potent inhibitor of GABA reuptake, has not been associated with any teratogenic effects in experimental models.

Unlike carbamazepine, oxcarbazepine is not metabolised into the 10–11 epoxide metabolite; however, it is not known whether this results in lower teratogenicity. In women receiving oxcarbazepine, 23 pregnancies have been reported: 19 women gave birth to healthy offspring, three had spontaneous abortions and one case of spina bifida was diagnosed.(8,9)

Zonisamide blocks voltage-sensitive sodium channels and voltage-dependent calcium currents. Of 26 reported pregnancies, birth defects were reported in two with exposure to zonisamide in combination.

Only nine pregnancies in women treated with felbamate have been reported, and no birth defects were detected. The number of pregnancies with ­topiramate exposure is not known.

In order to understand the risks of AEDs in pregnancy, an international pregnancy registry has been established (EURAP). The primary objective of this registry is to evaluate and determine the comparative degree of safety of AEDs in the human fetus, with reference to new and old AEDs and individual drugs and drugs in combinations.

Coadministration of folic acid and AEDs is highly recommended in all women with epilepsy of childbearing age in order to prevent congenital malformations, particularly neural tube defects.(10,11) There is no consensus regarding the optimal dosage of folic acid. The lowest recommended daily dosage should be 0.4mg. The daily dosage of folic acid in patients undergoing treatment with AEDs varies between
2 and 4mg. In patients with a family history of neural tube defects or receiving valproate or in polytherapy, dosages of 4mg of folic acid are suggested.

Oral vitamin K should be given to pregnant women receiving enzyme-inducing AEDs (phenytoin, carbamazepine, valproate and primidone).(12–14) These AEDS are competitive inhibitors of prothrombin precursors and may lead to neonatal haemorrhage. The risk of haemorrhagic disease of the newborn appears to be less than 7% and extends at least one week into the puerperium. The risk is higher in the first 24 hours after birth. The recommended daily dosage for the mother is 10mg during the last month of pregnancy. The infant should be injected with vitamin K immediately after delivery.

Prenatal diagnosis with a-fetoprotein (AFP) should be offered at 14–16 weeks’ gestation, and structural ultrasound at 16–20 weeks’. If indicated, amniocentesis and acetylcholinesterase levels should be recommended.(1,15)

Oral AEDs should be taken during labour. If vomiting is present a repeat oral dose in an hour is recommended. Hyperventilation or exhaustion should be avoided. Epidural anaesthesia can be safely performed. Convulsive seizures during labour and delivery should be treated with IV lorazepam. If seizures continue, administration of fosphenytoin is suggested and caesarean section should be considered. Caesarean section may be necessary in cases with poor seizure control towards the end of pregnancy.(16) Lorazepam is not recommended prophylactically, unless there is history of seizures during labour.

Status epilepticus is rare during pregnancy (two in 153 pregnancies).(17) It occurs during the last trimester and has been associated with significant maternal and fetal mortality. Management should include adequate oxygenation and control of acidaemia, both of which can compromise placental perfusion. Intravenous lorazepam, fosphenytoin or valproate should be given. A caesarean section should be considered.

AED withdrawal should be considered in some cases before conception. The American Academy of Neurology has issued guidelines for AED withdrawal in pregnancy.(18) Medication withdrawal should be considered only if the patient has been seizure-free for two years, has a normal neurological exam, normal brain MRI and no history of status epilepticus or prolonged seizure. If AED discontinuation is considered it should be completed at least six months before pregnancy. AEDs should not be changed during ­pregnancy due to increased risk of seizures.

Proper management of women with epilepsy should be instituted well before conception. Comprehensive care by a team including a neurologist, an obstetrician/gynaecologist and a paediatrician is advocated.

Strategies for management of pregnancy in women with epilepsy are listed in Table 2.



  1. Delgado-Escueta AV, Janz D. Neurology 1992;42 Suppl 5:149-60.
  2. Schmidt D, et al. J Neurol Neurosurg Psychiatry 1983;46:751-5.
  3. Yerby MS, et al. Epilepsy Res 1990;5:223-8.
  4. Steegers-Theunissen R, et al. Epilepsy Res 1994;18:261-9.
  5. Koch S, et al. Neurology1992;42 Suppl 5:83-8.
  6. Holmes B L, et al. N Engl J Med 2001;344:1132-8.
  7. Ohman I, et al Epilepsia 2000;41(6):709-13.
  8. Fris ML, et al. Acta Neurol Scand 1993;87:224-7.
  9. Lindout D, Omtzigt JGC. Epilepsia 1994;35 Suppl 4:S19-28.
  10. Lewis DP, et al. Ann Pharmacother 1998;32:947-1095.
  11. Dansky LV, et al. Neurology 1992;42 Suppl 5:32-42.
  12. Bleyer WA, Skinner AL. JAMA 1976;235:626-7.
  13. Mountain KR, et al. Lancet 1970;1:265-8.
  14. Cornelissen M, et al. Am J Obstet Gynecol 1993a;168:923-8.
  15. Hobbins JC. N Engl J Med 1991;324:690-1.
  16. Betts T, Crawford P. Women and epilepsy. London: Martin Dunitz; 1998. p. 40-1.
  17. Teramo K, Hiilesmaa V. In: Epilepsy, pregnancy and the child. Janz D, ed. New York: Raven Press; 1982. p. 53-9.
  18. Quality Standards Subcommittee of the American Academy of Neurology. Neurology 1998;51:944-8.
  19. Zahn C. Epilepsia1998;39 Suppl 8:S26-31.
  20. Tran A, et al. Br J Clin Pharmacol 1998;45:409-11.
  21. Shimoyama R, et al. Biomed Chromatogr 1999;13(5):370-2.

5th European Congress on Epileptology
6–10 October 2002
Madrid, Spain
25th International Epilepsy Congress
12–16 October 2003
Tunis, Tunisia

Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine