In three to five of 1000 births, the mother has epilepsy . With the use of antiepileptic drugs (AEDs) for other indications, the proportion of women taking AEDs during pregnancy and lactation is substantial. In this situation the risks associated with drug exposure to the fetus and newborn need to be balanced against the risks incurred by seizures, and knowledge of pharmacokinetic alterations becomes particularly important for AED optimization.
This review provides an overview of the pharmacokinetics of AEDs in pregnancy and lactation and possible clinical consequences (Table 1). Related aspects, for example teratogenic risks and vitamin supplementation, are not covered.
Maternal AED plasma concentrations are important as they are related to the antiepileptic effects in the treated woman as well as reflecting the drug exposure to the fetus. The use of AEDs during pregnancy has been associated with increased rates of birth defects and neuro-cognitive deficits [2–4,5••] and these risks seem to be related to the number of AEDs, drug dosages and thus also presumably to drug concentrations [6–8]. Given the importance of drug exposure during the first gestational weeks, appropriate prepregnancy counseling is critical. Therefore, and because many pregnancies are unplanned, the possibility of a future pregnancy always has to be considered when prescribing AEDs to women of childbearing potential. The information has to be repeated regularly and needs to be qualified and balanced to avoid the risk of abrupt drug withdrawal because of maternal fear of adverse effects of the medication on the developing fetus .
Although recent studies indicate differences between AEDs in teratogenic potential [10,11•], it is beyond the scope of this review to discuss drug selection for women with epilepsy who plan to become pregnant. However, if pregnancy is planned in advance, this is the appropriate time to reassess AED treatment. For those with active epilepsy, treatment should aim at seizure control (in particular of generalized tonic–clonic seizures) with monotherapy at the lowest effective dose. This should be tried out and assessed before pregnancy, and when it is achieved it is of value to determine the optimal AED plasma concentration in the individual patient before pregnancy. This can be used as reference to help in the interpretation of drug levels obtained during pregnancy [12•].
Estradiol containing oral contraceptives induce the metabolism of lamotrigine  and may have similar effects on valproate . Therefore, it is advisable to also monitor plasma concentrations of these AEDs when a woman plans to discontinue hormonal oral contraceptives to become pregnant. The AED dosage may need to be reduced to avoid unnecessarily high drug exposure to the fetus at the time of conception.
Pharmacokinetic alterations of antiepileptic drugs in pregnancy
Pregnancy can affect the pharmacokinetics of AEDs at all levels, absorption, distribution, metabolism and elimination , resulting in declining plasma concentrations of AEDs as pregnancy progresses. Increased renal elimination and especially increased metabolic clearance due to enzyme induction are the clinically most important mechanisms. The latter appears to be particularly marked for AEDs that are cleared by glucuronidation [16•,17]. Decreased binding to plasma proteins may contribute to declining total plasma concentrations of AEDs that are highly protein bound, for example, phenytoin, valproate and tiagabine. However, although decreased binding will result in lower total drug levels (bound plus unbound), the unbound and pharmacologically active concentration of the drug remains relatively unchanged. As the unbound concentration also reflects the drug exposure of the fetus, is it recommended, if possible, to monitor the free concentrations instead of the total concentrations of the highly protein bound drugs to avoid unnecessary dose increase. This is particularly important for valproate, which is associated with the most pronounced teratogenic potential .
The plasma concentrations tend to decrease in pregnancy for many of the AEDs and these alterations have to be taken into consideration to minimize the risk of seizure deterioration as pregnancy progress (for review see [18,19•]). Although the effects of pregnancy on drug disposition depend on the pharmacokinetic properties of the AED, the extent of these effects also varies between patients. The effects of pregnancy on plasma concentrations of different AEDs are summarized in Table 2.
Total concentrations of phenobarbital decline on average by 50%, of phenytoin by 50–60%, and of carbamazepine by up to 25%. The decline in unbound concentrations of carbamazepine is even smaller, and of phenytoin 20–30% [21–24]. Although valproate is predominantly eliminated by glucuronidation, unbound plasma concentrations are changed only slightly whereas total plasma concentrations of valproate could decrease by up to 50% by the end of the third trimester [25,26].
Among the newer generation AEDs, it is only lamotrigine that has been investigated in detail. Lamotrigine undergoes hepatic glucuronidation and this process is induced by pregnancy [16•]. The clearance of lamotrigine increases progressively until the 32nd gestational week when it may be two to three times higher than before pregnancy [27–31]. After delivery the lamotrigine elimination rate drops rapidly, within days, to reach the prepregnant level within the first 2–3 weeks postpartum. The extent to which pregnancy affects lamotrigine kinetics varies considerably between patients and white women have been reported to have higher plasma clearance during pregnancy than black women [12•]. Comedication also clearly has an influence. The effects of pregnancy on lamotrigine plasma concentrations are minor in women on comedication with valproate, a drug known to inhibit glucuronidation .
The active metabolite of oxcarbazepine, the monohydroxy metabolite (MHD) derivative, is also mainly metabolized by glucuronide conjugation . The clearance of MHD is, thus, also increased substantially during pregnancy. Plasma concentrations of MHD decline by 36 to 50% in the last trimester [17,33], which is slightly less than with lamotrigine, but more data are still needed for better understanding of oxcarbazepine kinetics.
Levetiracetam is mainly eliminated renally, but approximately 25% of the dose is metabolized by enzymatic hydrolysis . Recent data have demonstrated that the plasma level of levetiracetam is reduced to about 50% in the third trimester and increase to prepregnant levels within the first 2 weeks postpartum [35,36•]. The mechanism for this effect of pregnancy remains to be clarified.
No systematic studies have assessed the gestational influence on the pharmacokinetics of other newer generation AEDs. However, at delivery maternal plasma concentrations of topiramate and gabapentin were not markedly different from concentrations 2–3 weeks postpartum in the three reported cases on each drug. Two case reports suggest slightly lower maternal plasma concentrations of zonisamide at delivery compared with the first 2 weeks postpartum [39,40•].
Several other new AEDs, for example, pregabalin, tiagabine and vigabatrin have been marketed within the last 10 to 15 years, but no published data on the pharmacokinetics during pregnancy exist for these drugs. Both pregabalin and vigabatrin have negligible protein binding and are eliminated unchanged renally. Any change in serum concentrations of these drugs can be expected to mainly reflect the increase in renal glomerular filtration. Tiagabine, on the contrary, is highly protein bound (>90%) and extensively metabolized. Therefore, considerable gestational changes can be expected. Because of the effect of displacement of binding sites of tiagabine as for other high protein bound drugs, it is likely that a decrease in total tiagabine concentration is more pronounced than the drop in the active unbound concentration.
Seizure control and changes in antiepileptic drug levels in pregnancy
Those with well controlled epilepsy are probably overrepresented among women with epilepsy that become pregnant. In this population, even a single breakthrough seizure induced by pharmacokinetic alterations may pose serious consequences for everyday life. In addition, poorly controlled tonic–clonic seizures during pregnancy may adversely affect maternal health as well as fetal development .
The largest prospective single study, including 1736 pregnancies, reported that 59% remained seizure free throughout pregnancy . Other studies have indicated that the risk of seizures during pregnancy is less if the prepregnancy year was seizure free [43•]. However, the risk of seizure deterioration during pregnancy varies widely among different studies. In a review of data published since 1980, covering more than 4000 pregnancies, an increase in seizures was found in on average 25% of pregnancies (range 13–47%) [44••]. Seizure control was unchanged from before pregnancy in on average 57% of the women.
In the large prospective European Epilepsy and Pregnancy Registry study, treatment with oxcarbazepine was associated with an increased risk of seizures, and the need for dose adjustments was significantly related to treatment with oxcarbazepine and lamotrigine . The Australian Pregnancy Register reported that the control of convulsive seizures was significantly worse with lamotrigine compared with valproate or carbamazepine . These observations could be a reflection of the fact that lamotrigine and oxcarbazepine are the two AEDs with the most pronounced gestational-related pharmacokinetic changes. In fact, the drop in lamotrigine plasma concentrations in pregnancy has frequently been associated with an increase in seizures [30,31]. A recent study of 53 pregnancies with lamotrigine treatment clearly demonstrated the correlation between declining drug concentrations and increase in seizures [12•]. A decrease in lamotrigine plasma concentration from prepregnancy levels by 35% or more was a significant predictor of seizure deterioration.
Although the association between pharmacokinetic alterations and seizure control has been most convincingly demonstrated for lamotrigine, it is reasonable to assume that for AEDs in general declining active drug concentrations in pregnancy indicate an increased risk of seizures. This is the justification for regular drug level monitoring during pregnancy of AEDs where alterations can be expected but their magnitude is difficult to predict. In settings where therapeutic drug monitoring is unavailable, it might sometimes be advisable to double the dose of lamotrigine and increase the dose of oxcarbazepine/levetiracetam gradually from the second trimester.
The safety of antiepileptic drugs and breastfeeding
Women with epilepsy are in general encouraged to breastfeed their infants as the benefits of breastfeeding are considered to outweigh the potential adverse effects of continued AED exposure for the nursed infant . The amount of AED exposure for a breast-fed infant depends on the maternal plasma concentration of the drug, the fraction of the AED transferred to the breast milk and on the infant absorption and elimination capacity. AEDs are transferred into breast milk in an inverse proportion to their extent of protein binding, and AEDs with low protein binding, low molecular weight and high lipophilicity are most likely to accumulate in breast milk  (Table 2) (for review see ). Phenytoin, valproate and tiagabine are more than 90% protein bound and, therefore, have low concentrations in breast milk, whereas the amount of carbamazepine, phenobarbital, oxcarbazepine, lamotrigine, topiramate and zonisamide that pass into breast milk are low to moderate. Ethosuximide, vigabatrin, gabapentin, pregabalin and levetiracetam have no protein binding and it should, therefore, be expected that drug passage into breast milk is extensive and the concentration equivalent to the maternal plasma level. This has been confirmed in studies on gabapentin and levetiracetam [35,36•,47]. The plasma concentrations in the breast-fed infants, however, are low in each case, indicating an efficient elimination in newborns.
For neonatal or prenatal infants, there is a risk that the capacity of the immature drug metabolizing pathways is low and medications may be accumulated at concentrations with a pharmacological potential. This has been demonstrated for barbiturates with a risk of sedation of the infant  and may also be applicable to AEDs that are eliminated by glucuronidation, for example, lamotrigine and the monohydroxy derivative of oxcarbazepine. Infant plasma concentrations of lamotrigine up to 40% of the maternal plasma concentration have been reported . In addition, the free fraction seems to be higher in the plasma of the nursing infant than in the mother. A recent study demonstrated that the infant/maternal plasma concentration ratio for unbound lamotrigine was 30.9% compared with 18.3% for total lamotrigine [49•]. The same study demonstrated a considerable interpersonal variability in the ratio of infant/maternal plasma concentration. However, no study has reported clinically relevant adverse effects in lamotrigine exposed breast-fed infants. Nor did the only available case report of oxcarbazepine and breastfeeding indicate any adverse effects in the suckling infant .
The effect of pregnancy on the pharmacokinetics of AEDs is difficult to predict. Drug levels should, therefore, preferably be monitored regularly throughout pregnancy, in particular if the woman is treated with AEDs known to display pronounced changes in plasma concentrations. Monitoring should be tailored to the clinical characteristics of the individual patient. It is evident that there are important gaps in the current knowledge and it is essential to establish studies which prospectively and systematically explore this area.
No funding has been received for this work. Anne Sabers has received speaker's fee and/or research grants from the manufacturers of levetiracetam (UCB Pharma) and rufinamide and zonisamide (Eisai). Torbjörn Tomson received speaker's fees and/or research grants from the manufacturers of carbamazepine and oxcarbazepine (Novartis), carisbamate and topiramate (Johnson and Johnson), ethosuximide, gabapentin, phenytoin and pregabalin (Pfizer), lamotrigine (GSK), lacosamide and levetiracetam (UCB Pharma), tiagabine, valproic acid and vigabatrin (Sanofi-Aventis), rufinamide and zonisamide (Eisai).
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Additional references related to this topic can also be found in the Current World Literature section in this issue (pp. 000–000).
1 Hauser WA, Annegers JF, Rocca WA. Descriptive epidemiology of epilepsy: contributions of population-based studies from Rochester, Minnesota. Mayo Clin Proc 1996; 71:576–586.
2 Fried S, Kozer E, Nulman I, et al
. Malformation rates in children of women with untreated epilepsy. Drug Safety 2004; 27:197–202.
3 Gaily E, Kantola-Sorsa E, Hiilesmaa V, et al
. Normal intelligence in children with prenatal exposure to carbamazepine. Neurology 2004; 62:28–32.
4 Tomson T, Hiilesmaa V. Epilepsy in pregnancy
. BMJ 2007; 335:769–773.
5•• Meador KJ, Pennell PB, Harden CL, et al
registries in epilepsy: a consensus statement on health outcomes. Neurology 2008; 71:1109–1117. Consensus report describing the current state of knowledge in antiepileptic drug treatment during pregnancy
6 Samren EB, van Duijn CM, Koch S, et al
. Maternal use of antiepileptic drugs and the risk of major congenital malformations: a joint European prospective study of human teratogenesis associated with maternal epilepsy. Epilepsia 1997; 38:981–990.
7 Kaneko S, Battino D, Andermann E, et al
. Congenital malformations due to antiepileptic drugs. Epilepsy Res 1999; 33:145–158.
8 Adab N. Birth defects and epilepsy medication. Expert Rev Neurother 2006; 6:833–845.
9 Williams J, Myson V, Steward S, et al
. Self-discontinuation of antiepileptic medication in pregnancy
: detection by hair analysis. Epilepsia 2002; 43:824–831.
10 Morrow J, Russell A, Guthrie E, et al
. Malformation risks of antiepileptic drugs in pregnancy
: a prospective study from the UK Epilepsy and Pregnancy
Register. J Neurol Neurosurg Psychiatry 2006; 77:193–198.
11• Meador K, Reynolds MV, Crean S, et al
outcomes in women with epilepsy: a systematic review and metaanalysis of published pregnancy
registries and cohorts. Epilepsy Res 2008; 81:1–13. Comprehensive systematic review of 59 studies evaluating the risk of fetal malformations in children exposed to antiepileptic drugs in utero.
12• Pennel PB, Peng L, Newport DJ, et al
. Lamotrigine in pregnancy
. Clearance, therapeutic drug monitoring, and seizure frequency. Neurology 2008; 70:2130–2136. This study describes the alterations of total and free lamotrigine plasma clearance during pregnancy
and postpartum and contributes to clarification of the rationale for a dosing paradigm for lamotrigine in pregnancy
13 Sabers A, Öhman I, Christensen J, Tomson T. Oral contraceptives reduce lamotrigine plasma levels. Neurology 2003; 61:570–571.
14 Galimberti CA, Mazzucchelli I, Arbazino C, et al
. Increased apparent oral clearance of valproic acid during intake of combined contraceptive steroids in women with epilepsy. Epilepsia 2006; 47:1569–1572.
15 Pennel PB. Antiepileptic drug pharmacokinetics during pregnancy
and lactation. Neurology 2003; 61(Suppl 2):35–42.
16• Öhman I, Luef G, Tomson T. Effects of pregnancy
and contraception on lamotrigine disposition: new insight through analysis of lamotrigine metabolites. Seizure 2008; 17:199–202. This study confirms that the pharmacological background for increased lamotrigine metabolism in pregnancy
is caused by increased glucuronidation processes.
17 Mazzucchelli I, Onat FY, Ozkara C, et al
. Changes in the disposition of oxcarbazepine and its metabolites during pregnancy
and the puerperium. Epilepsia 2006; 47:504–509.
18 Tomson T, Battino D. Pharmacokinetics and therapeutic drug monitoring of newer antiepileptic drugs during pregnancy
and the puerperium. Clin Pharmacokinet 2007; 46:209–219.
19• Sabers A, Petrenaite V. Pharmacokinetics of antiepileptic drugs in pregnancy
. Expert Rev Clin Pharmacol 2008; 1:129–136. A review of pharmacokinetic alterations of specific antiepileptic drugs in pregnancy
20 Tomson T. Gender aspects of pharmacokinetics of new and old AEDs. Pregnancy
and breastfeeding. Ther Drug Monit 2005; 27:718–721.
21 Bardy AH, Teramo K, Hiilesmaa VK. Apparent clearance of phenytoin, phenobarbitone, primidone, and carbamazepine during pregnancy
: results of the prospective Helsinki study. In: Janz D, Dam M, Richens A, Bossi L, Helge H, Schmidt D, editors. Epilepsy, pregnancy
, and the child. New York: Raven Press; 1982. pp. 141–145.
22 Yerby MS, Friel PN, McCormick K, et al
. Pharmacokinetics of anticonvulsants in pregnancy
: alterations in plasma protein binding. Epilepsy Res 1990; 5:223–228.
23 Lander CM, Eadie MJ. Plasma antiepileptic drug concentration during pregnancy
. Epilepsia 1991; 32:257–266.
24 Tomson T, Lindbom U, Ekqvist B, Sundqvist A. Disposition of carbamazepine and phenytoin in pregnancy
. Epilepsia 1994; 35:131–135.
25 Yerby MS, Friel PN, McCormick K. Antiepileptic drug disposition during pregnancy
. Neurology 1992; 42(Suppl 5):12–16.
26 Philbert A, Petersen B, Dam M. Concentration of valproate during pregnancy
, in the newborn and in breast milk. Acta Neurol Scand 1985; 72:460–463.
27 Öhman I, Vitols S, Tomson T. Lamotrigine in pregnancy
: pharmacokinetics during delivery, in the neonate and during lactation. Epilepsia 2000; 41:709–713.
28 Pennel PB, Newport DJ, Stowe ZN, et al
. The impact of pregnancy
and childbirth on metabolism of lamotrigine. Neurology 2004; 62:292–295.
29 Tran TA, Leppik IE, Blesi K, et al
. Lamotrigine clearance during pregnancy
. Neurology 2002; 59:251–255.
30 De Haan GJ, Edelbroek P, Segers J, et al
. Gestation-induced changes in lamotrigine pharmacokinetics: a monotherapy study. Neurology 2004; 63:571–573.
31 Petrenaite V, Sabers A, Hansen-Schwartz J. Individual changes in lamotrigine plasma concentrations during pregnancy
. Epilepsy Res 2005; 65:185–188.
32 Tomson T, Luef G, Sabers A, et al
. Valproate effects on kinetics of lamotrigine in pregnancy
and treatment with oral contraceptives. Neurology 2006; 67:1297–1299.
33 Christensen J, Sabers A, Sidenius P. Oxcarbazepine concentrations during pregnancy
: a retrospective study in patients with epilepsy. Neurology 2006; 67:1497–1499.
34 Patsalos PN. Pharmacokinetic profile of levetiracetam: toward ideal characteristics. Pharmacol Ther 2000; 85:77–85.
35 Tomson T, Palm R, Källén K, et al
. Pharmacokinetics of levetiracetam during pregnancy
, delivery, in the neonatal period and lactation. Epilepsia 2007; 48:1111–1116.
36• Westin AA, Reimers A, Helde G, et al
. Serum concentration/dose ratio of levetiracetam before, during and after pregnancy
. Seizure 2008; 17:192–198. This study confirms retrospectively the previous observation that the elimination of levetiracetam is accelerated in pregnancy
37 Öhman I, Vitols S, Luef G, et al
. Topiramate kinetics during delivery, lactation, and in the neonate: preliminary observations. Epilepsia 2002; 43:1157–1160.
38 Öhman I, Vitols S, Tomson T. Pharmacokinetics of gabapentin during delivery, in the neonatal period and lactation: does a fetal accumulation occur during pregnancy
. Epilepsia 2005; 46:1621–1624.
39 Kawada K, Itoh S, Kusaka T, et al
. Pharmacokinetics of zonisamide in perinatal period. Brain Dev 2002; 24:95–97.
40• Oles KS, Bell WL. Zonisamide concentrations during pregnancy
. Ann Pharmacother 2008; 42:1139–1141. This case-based paper shows that the plasma concentration of zonisamide is decreased at the end of the second trimester.
41 Adab N, Kini U, Vinten J, et al
. The longer term outcome of children born to mothers with epilepsy. J Neurol Neurosurg Psychiatry 2004; 75:1575–1583.
42 EURAP study group. Seizure control and treatment in pregnancy
: observations from the EURAP Epilepsy Pregnancy
Registry. Neurology 2006; 66: 354–360.
43• Vajda FJE, Hitchcock A, Graham J, et al
. Seizure control in antiepileptic drug-treated pregnancy
. Epilepsia 2008; 49:172–175. This study covering 841 antiepileptic drug-treated pregnancies shows that the risk of seizures during pregnancy
is significantly less if the patient has been seizure-free 1 year prior to pregnancy
44•• Battino D, Tomson T. Seizure control in pregnancy
. In: Panayiotopoulos CP, Crawford P, Tomson T, editors. Epilepsies in girls and women. UK: MEDICINAE, Oxford; 2008. pp. 138–142.
This paper describes alterations of seizure frequency during pregnancy
on the basis of a review of studies published since 1980 covering more than 4000 pregnancies.
45 Vajda FJE, Hitchcock A, Graham J, et al
. Foetal malformations and seizure control: 52 months data of the Australian Pregnancy
registry. Eur J Neurol 2006; 13:645–654.
46 Practice parameter management issues for women with epilepsy (summary statement). Report of the quality standards subcommittee of the American Academy of Neurology. Neurology 1998; 51: 944–948.
47 Johannessen SI, Helde G, Brodtkorb E. Levetiracetam in serum and in breastmilk at birth and during lactation. Epilepsia 2005; 46:775–777.
48 Kuhnz W, Kock S, Helge H, et al
. Primidone and phenobarbital during lactation period in epileptic women: total and free drug serum levels in the nursed infants and their effects on neonatal behaviour. Dev Pharmacol Ther 1988; 11:147–154.
49• Newport DJ, Pennell PB, Calamares MR, et al
. Lamotrigine in breast milk and nursing infants: Determination of exposure. Pediatrics 2008; 122:223–231. In this study it is shown that the excretion of lamotrigine into breastmilk is highly variable. The mean milk/maternal plasma ratio, however, is not higher than described for other antiepileptic drugs.
50 Bülau P, Paar WD, van Unruh GE. Pharmacokinetics of oxcarbazepine and 10-hydroxy-carbazepine in the newborn child of an oxcarbazepine-treated mother. Eur J Clin Pharmacol 1988; 34:311–313.