INTRODUCTION — While most pregnancies in women with epilepsy are uncomplicated, the risk for a range of perinatal complications is increased compared with the general population, including preeclampsia, hemorrhage, fetal growth restriction, stillbirth, and maternal mortality [1-4]. Offspring are at increased risk for major congenital malformations and adverse neurodevelopmental outcomes. The degree of maternal metabolic alteration associated with antiseizure medications (ASMs) is increasingly appreciated and may represent a pathologic intermediary to clinically apparent fetal and maternal affects [5]. However, many of these risks can be modified by careful selection of the type and dose of ASMs prior to conception and continuing throughout pregnancy [6-8].
This topic will discuss the risks associated with epilepsy and pregnancy. The management of pregnancy in women with epilepsy is discussed separately. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period".)
PERINATAL MORBIDITY AND MORTALITY — A growing body of literature indicates that perinatal morbidity and mortality are increased among women with epilepsy compared with the general population; complications range from mild to severe and include preeclampsia, preterm labor, bleeding, placental abruption, poor fetal growth, prematurity, fetal death, and maternal mortality [1,3,9-16]. The magnitude of the increase in risk appears to be relatively small for most complications (between 1 and 1.7 times expected rates), with the exception of maternal mortality, which may be as much as 10-fold increased [8]. However, this translates to a very small increase in the absolute risk of maternal mortality.
Maternal mortality — Several studies have found that mortality rates during pregnancy are approximately 10-fold higher in pregnant women with epilepsy compared with the general population of pregnant women [3,4,17]. This translates to a low absolute increase in risk of less than 0.1 percent.
There are a number of possible explanations for this finding, including an increase in medical comorbidities, an increase in life-threatening complications of pregnancy, and an increase in seizure-related complications, including sudden unexpected death in epilepsy (SUDEP). In one study of epilepsy-related deaths in a United Kingdom population-based registry, 11 out of 14 deaths (79 percent) were attributed to SUDEP; approximately one-third occurred during the postpartum period [17]. Two-thirds of the deaths were in patients treated with lamotrigine (a possible risk factor for SUDEP in females), but it is not clear if this was merely a reflection of prescribing practices or could be related to the substantial lowering of lamotrigine concentrations that occur during pregnancy if doses are not adjusted (see "Management of epilepsy during preconception, pregnancy, and the postpartum period"). SUDEP is discussed in more detail separately. (See "Sudden unexpected death in epilepsy".)
Fetal death or stillbirth — There may be a modest increase in the risk of fetal death or stillbirth among offspring of individuals with epilepsy, although studies are not entirely consistent. Both a 2015 systematic review and meta-analysis [1] and a large population-based retrospective cohort study of delivery in United States hospitals between 2007 and 2011 [3] found a small increased risk of miscarriage or stillbirth. In the US study, the risk of stillbirth was higher for women with epilepsy compared with women without epilepsy (0.8 versus 0.6 percent, adjusted OR 1.27, 95% CI 1.17-1.38) but the absolute increase in risk was only 0.2 percent [3]. However, in a prospective observational study at four United States academic medical centers, miscarriage rates in the women with epilepsy were the same as in control women [18].
The mechanism of an increased risk of fetal death among pregnant women with epilepsy is also not well understood. In a large European registry, only 1 of 165 reported miscarriages or stillbirths was associated with seizures or status epilepticus; two-thirds of these pregnancies were seizure-free [19]. Whether ASM exposure contributes to risk of spontaneous miscarriage is uncertain, but in aggregate the contribution appears to be minimal [1]. In a large population-based observational study that included nearly one million pregnancies, use of ASMs during pregnancy conferred a slightly higher risk of spontaneous abortion compared with not using ASMs (adjusted risk ratio [aRR] 1.13, 95% CI 1.04-1.24); however, when limited to women with an epilepsy diagnosis, the use of ASMs was not associated with increased risk (aRR 0.98, 95% CI 0.87-1.09) [20].
Additionally, there appears to be a great deal of geographic variation in pregnancy outcomes [21]. Paradoxically, the risks of complications of hemorrhage and stillbirth appeared highest in high-income countries.
Preterm birth — ASM exposure is associated with an increased risk of preterm birth [22]. Interestingly, the effect was also present among females prescribed ASMs for a psychiatric indication, at least suggesting that the effect may be medication-associated. These results are similar to those of an earlier national registry of 679,762 Danish singleton births, of which 2982 were exposed to ASMs [23]. In this cohort, however, the risk of prematurity was limited to those without a history of epilepsy.
Other outcomes — In addition to fetal death, stillbirth, and preterm birth, epilepsy is associated with a small but significant increase in risk for a number of additional perinatal outcomes. In a large study of delivery hospitalizations from United States hospitals between 2007 and 2011, risk was moderately higher among women with epilepsy compared with women without epilepsy, for the following outcomes [3]:
●Cesarean delivery: 41 versus 33 percent (odds ratio [OR] 1.4, 95% CI 1.38-1.42)
●Pregnancy-related hypertension (10.5 versus 7.9 percent, OR 1.3, 95% CI 1.27-1.33)
●Preeclampsia (6.7 versus 4.2 percent, OR 1.59, 95% CI 1.54-1.63)
●Antepartum hemorrhage (2.1 versus 1.5 percent, OR 1.38, 95% CI 1.31-1.45)
●Postpartum hemorrhage, including severe postpartum hemorrhage (0.7 versus 0.4 percent, OR 1.76, 95% CI 1.61-1.93)
●Preterm labor (11 versus 7 percent; OR 1.54, 95% CI 1.50-1.57)
●Poor fetal growth (3.7 versus 2.1 percent, OR 1.68, 95% CI 1.61-1.75)
Effect of seizures on the fetus — In addition to concerns about fetal exposure to ASMs, there are risks to the fetus from maternal seizures. In particular, generalized tonic-clonic seizures can lead to hypoxia and lactic acidosis, which may harm the fetus via placental transfer [8]. Although probably less harmful, other types of seizures may increase the risk of fetal growth restriction, fetal injury, and premature delivery. However, few studies have evaluated the direct effects of maternal seizures on the fetus.
Fetal hypoxia may occur as a result of maternal hypoxia, decreased placental blood flow, or postictal apnea, but no information is available on the number or length of seizures that may jeopardize the fetus. One report of fetal heart rate monitoring during a maternal generalized tonic-clonic seizure lasting 2.5 minutes revealed significant fetal heart rate deceleration lasting up to 30 minutes after the seizure [24]. While nonconvulsive seizures are believed to be less dangerous [8], another case report documented significant fetal bradycardia during a one-minute, focal impaired awareness seizure (also called complex partial seizure) [25].
Additional risks of maternal seizures include injury to the fetus, placental abruption, or fetal loss due to maternal trauma sustained during a seizure.
In a population-based data set in Taiwan, compared with women with epilepsy who did not suffer seizures during pregnancy, epileptic seizures during pregnancy were marginally associated with small for gestational age (SGA) infants (adjusted odds ratio [OR] 1.34, 95% CI 1.01-1.84) but not with preterm birth (OR 1.12, 95% CI 0.73-1.71) [26]. A study of 249 children in the United Kingdom found that the number of generalized tonic-clonic convulsive seizures during pregnancy was a negative predictor of verbal intelligence quotient (IQ) in the children; in particular, the occurrence of five or more convulsive seizures during pregnancy was associated with developmental delay [27].
In a large European registry of 1956 pregnancies in 1882 women with epilepsy, status epilepticus (SE) occurred in 1.8 percent, one-third of which were convulsive type [19]. Only one case of SE was associated with stillbirth; the remaining cases were not clearly associated with fetal or maternal complications.
EFFECT OF PREGNANCY ON SEIZURES
Seizure control — The effect of pregnancy and antiseizure medications (ASM) management on seizure control is reviewed in the following sections.
●Evidence of seizure worsening during pregnancy – Although there is variability in the published literature, most earlier studies reported that 20 to 50 percent of women have worsening of seizures during pregnancy compared with their baseline [12,28-33]. The 2006 International Registry of Antiepileptic Drugs and Pregnancy (EURAP) also reported on seizure control in pregnancy, but the comparator was the first trimester rather than the nonpregnant baseline [19]. Among 1956 pregnancies in 1882 women with epilepsy, seizure frequency and ASM treatment remained unchanged in 62 to 64 percent; 59 percent were seizure-free during pregnancy; only 17 percent reported increased seizure frequency. Status epilepticus (SE) occurred in approximately 2 percent, one-third of which were convulsive type. A 2019 systematic review of the literature from the International League Against Epilepsy (ILAE) Task Force on Women and Pregnancy concluded that approximately two-thirds of women with epilepsy maintain baseline seizure control during pregnancy [8]. However, none of these studies compared seizure frequency among pregnant women to the gold standard of seizure frequency among nonpregnant women with epilepsy followed over the same length of time and in the same manner; therefore, the evidence was insufficient to determine the change in seizure frequency for pregnant women with epilepsy [12].
●Prevention of seizure worsening during pregnancy – Meticulous ASM management in conjunction with therapeutic drug monitoring during pregnancy may prevent the increased risk of seizures during pregnancy. Supporting evidence is provided by a 2020 prospective observational study from the Maternal Outcomes and Neurodevelopmental Effects of Antiepileptic Drugs (MONEAD) Study Group, which compared the proportion of women with increased seizure frequency during pregnancy with a matched, nonpregnant control group followed for the same length of time and with the same protocol [34]. The main analysis included 299 pregnant women and 93 controls who had a history of seizures that impaired awareness. Seizure frequency was higher during pregnancy than the individual woman's nongravid baseline in 23 percent of pregnant women compared with 25 percent of controls during the corresponding time epochs (odds ratio [OR] 0.93, 95% CI 0.54-1.60). However, in this observational study, the two groups differed in medication management. The ASM dose was changed at least once in 74 percent of the pregnant women compared with only 31 percent of the controls (OR 6.36, 95% CI 3.82-10.59). Moreover, doses of individual ASMs were increased during pregnancy in alignment with reported alterations in the clearance of that ASM during pregnancy, and doses were decreased again after delivery. These findings will help clinicians to counsel women with epilepsy on expectations for seizure control during pregnancy.
●Seizure control with catamenial epilepsy – Women with catamenial epilepsy may have improved seizure control during pregnancy, possibly attributable to the absence of cyclical hormone variations and increased progesterone and allopregnanolone levels. In a prospective study of 274 women with epilepsy, the women with a history of a catamenial pattern (n = 59) were more likely to be seizure free or experience seizure reduction compared with their nonpregnant baseline, with odds ratios (ORs) in the 2.42 to 2.62 range compared with the women without a history of a catamenial pattern [35].
●Seizures during the peripartum period – Seizures may be more likely to occur during the peripartum period. A study of women enrolled in the Kerala Registry of Epilepsy and Pregnancy in India found that seizure relapse was the highest during the three peripartum days (one day prior to one day after the day of delivery) [33]. EURAP reported that seizures occurred during delivery in 3.5 percent of the enrolled women, and one woman had convulsive SE [19]. In the vast majority of the cases, however, labor and delivery are not associated with increased complications.
Predictors of seizure worsening — The main risk factor for seizures during pregnancy is baseline seizure frequency before pregnancy; women who are seizure-free in the nine months prior to pregnancy are less likely to have seizure worsening [8,19,33,34]. Other reported predictors are focal epilepsy syndromes [8,19,33], ASM type, polytherapy, patient adherence, and the use of therapeutic drug monitoring [12,28-33,36,37].
Some women who experience increased seizure frequency are sleep-deprived or nonadherent with their medications because of concerns about the effects of the medication on the developing fetus [8,38,39]. Altered ASM pharmacokinetics also contribute to increased seizure frequency during pregnancy. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period", section on 'Antiseizure medication monitoring and dose adjustment'.)
DEPRESSION AND ANXIETY — Individuals with epilepsy have an increased prevalence of comorbid depression and anxiety compared with individuals without epilepsy, and this appears to be true during pregnancy and the postpartum period as well.
In a population-based study that included 706 pregnancies in women with epilepsy and over 100,000 pregnancies in controls without epilepsy, peripartum depression affected 27 percent of those with epilepsy compared with 23 percent of those with other chronic diseases and 19 percent of the entire non-epilepsy population [40]. These rates were similar to those reported in a subsequent prospective single-center study [41]. Rates of anxiety were similarly elevated in women with epilepsy compared with all others (22 versus 15 percent) [40]. These studies suggest that risk factors for depression or anxiety among women with epilepsy include high seizure frequency, history of physical and/or sexual abuse, adverse socioeconomic factors, multiparity, previous loss of a child, antiseizure medication (ASM) use, in particular polytherapy, unplanned pregnancy, and prepregnancy depression or anxiety [40,41].
Detection of depression during pregnancy and postpartum is important because both pharmacologic and nonpharmacologic treatments are available, and untreated illness may have consequences for both mother and child. Several studies have suggested that females with epilepsy may be undertreated for depression compared with females without epilepsy [40,42], emphasizing the importance of screening and appropriate treatment. (See "Comorbidities and complications of epilepsy in adults", section on 'Depression and anxiety' and "Unipolar major depression during pregnancy: Epidemiology, clinical features, assessment, and diagnosis", section on 'Potential adverse outcomes' and "Mild to moderate episodes of antenatal unipolar major depression: Choosing treatment" and "Severe antenatal unipolar major depression: Choosing treatment".)
EFFECT OF ASMS ON THE FETUS AND NEONATE — One of the main adverse effects of antiseizure medications (ASMs) on the fetus and neonate is an increased risk of major congenital malformations. Risks associated with ASM use during pregnancy can be minimized by preconception planning and careful management during pregnancy. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period".)
Major congenital malformations and their risk factors — There is an increased risk of major congenital malformations in fetuses exposed to ASMs. The reported risk of major congenital malformations with maternal ASM use is 4 to 6 percent, compared with a population estimate of 2 to 3 percent [28,43-47]. A history of maternal epilepsy alone does not appear to increase the risk of fetal major congenital malformations [45].
●Types of malformations – The most common major malformations associated with ASMs are neural tube, congenital heart, and urinary tract defects, skeletal abnormalities, and oral clefts [43,48,49].
●Different ASMs have different risks – Different ASMs carry substantially different levels of risk for major malformations, with the risk of some ASMs very close to the general population and other ASMs (eg, valproate, phenytoin, phenobarbital, and topiramate) with several-fold increased risk for major malformations [50]. A 2019 report from the International League Against Epilepsy (ILAE) concluded that valproate is associated with the highest risk of major congenital malformations, while phenobarbital and topiramate are associated with an intermediate risk, and lamotrigine and levetiracetam are associated with the lowest risk [8]. The ILAE report also noted that the risk is dose-dependent for valproate and is probably dose-dependent for other ASMs including carbamazepine, phenobarbital and lamotrigine. The consistency of findings across different pregnancy studies and in varied regions of the world strengthen the conclusion that some ASMs are associated with a higher risk for major malformations than others (figure 1). (See 'Risks with specific ASMs' below and "Management of epilepsy during preconception, pregnancy, and the postpartum period", section on 'Changing antiseizure medications during pregnancy'.)
●Polytherapy – In general, ASM polytherapy is a risk factor for fetal major malformations as identified in a number of prospective registry studies, with the rates of major malformations increasing to 6 to 8.6 percent [45,47]. However, the risk is probably more dependent upon the specific ASMs used as polytherapy rather than just the number of ASMs [8]. Polytherapy with a regimen that includes valproate or topiramate represents a particularly high risk for major malformations [51-53]. Some polytherapy combinations do not have a substantially elevated risk for major malformations and are being used more commonly to avoid valproate use during pregnancy [54], especially in females of childbearing age with the epilepsy syndromes that are comprised of generalized-onset seizures.
●Timing and dose of ASM – In addition to the specific ASM used alone or in combination, the gestational timing of the exposure and the dose of the ASM used are also likely to be important. These have been best associated with valproate [46,49,55-57]. The International Registry of Antiepileptic Drugs and Pregnancy (EURAP) investigators analyzed the dose at the time of conception and reported a dose-dependent increase in the major malformation rates for monotherapy with valproate, phenobarbital, carbamazepine, and lamotrigine (figure 2) [49,50].
●History of malformation – A previous ASM-exposed pregnancy resulting in major malformations is associated with an increased risk in subsequent pregnancies [58,59]. The risk appears to be particularly high for valproate and topiramate but may also be increased for other ASMs and in pregnancies exposed to polypharmacy. In two different registry studies, women taking valproate who had a previous child with a major malformation had a recurrence risk of approximately 50 percent if they continued valproate during the subsequent pregnancy [58,59].
●Other risk factors – A family history of congenital anomalies, including from the paternal side, and a low maternal level of education have been identified as additional risk factors for fetal major malformations, at least in some studies [45,46,49,55,60-63]. Low folate levels are associated with higher risk for major malformations in pregnancy. Similar to the general population, folate supplementation of at least 0.4 mg per day is recommended for females with epilepsy at least one month prior to attempting conception and throughout the pregnancy [13]. (See "Folic acid supplementation in pregnancy", section on 'Dose and administration'.)
●Mechanism – The mechanism for ASM-induced teratogenicity has not been determined. One possibility is that some fetuses have low or deficient epoxide hydrolase activity that results in increased levels of teratogenic oxidative metabolites when they are exposed to ASMs [64]. Another mechanism involves oxidative damage to DNA from free radical intermediates produced by prostaglandin H synthase bioactivation of ASMs [65]. Deficiency of folic acid has also been implicated in teratogenicity associated with ASMs, some of which are folic acid antagonists. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period", section on 'Folic acid supplementation'.)
Fetal growth restriction — There is an increased risk of fetal growth restriction and delivery of a small for gestational age (SGA) infant with prenatal exposure to ASMs [8,22,23]. ASMs associated with increased risk in various studies include topiramate, zonisamide, carbamazepine, valproate, and phenobarbital [8,23,66]. The risk appears to be most pronounced with topiramate (see 'Topiramate' below). ASM polytherapy is also associated with an increased risk of fetal growth restriction [8,67].
Risks with specific ASMs — Across all pregnancy registries, valproate monotherapy is associated with the highest rates of major malformations [46,47,50]. Phenytoin and phenobarbital monotherapy have also been associated with relatively high baseline rates of major malformations [13,45,49,50,56,63,68-71]. Carbamazepine, topiramate, and zonisamide monotherapy are associated with modest rates of major malformations, while lamotrigine and levetiracetam and oxcarbazepine have lower rates major malformation that are similar to rates in the general population [49,72]. However, there are many additional ASMs with a paucity of data to allow even preliminary conclusions about the risk of major malformations with prenatal exposure. Decisions about selecting or changing ASMs with pregnancy are discussed in detail separately. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period".)
The spectrum of structural and neurodevelopmental teratogenic risks for different ASM monotherapies is depicted in the figure (figure 3). There are not enough data on the other ASMs to place them along the spectrum.
The risks of adverse effects on the fetus and neonate associated with individual ASMs are discussed in the sections that follow.
Valproate — Valproate should be avoided in pregnancy if possible. Valproate exposure in utero is associated with the development of neural tube-like defects (eg, spina bifida aperta, open lumbosacral myelocele) in 1 to 2 percent of fetuses, which represents a 10- to 20-fold increase over the general population [50,69,73,74]. Additional patterns of major malformations associated with first-trimester valproate exposure include oral clefts, as well as cardiovascular and urogenital malformations and multiple malformations [49,50,75,76].
First-trimester maternal exposure to valproic acid (VPA) increases the risk of major malformations, independent of any contribution of the epilepsy syndrome itself [10,13,77]:
●The North American Antiepileptic Drug Pregnancy Registry (NAAPR) followed 323 valproate-exposed pregnancies to completion [72,78]. The prevalence of major congenital anomalies in offspring of women who received valproate monotherapy was 9 percent, as compared with 3 percent in offspring of women receiving other ASMs and 1 percent in an unexposed control group. Thus, the relative risk of major malformations in VPA-exposed women was 9.0 (95% CI 3.4-23.3) [72]. Increased dose of VPA was also associated with increased risk (odds ratio [OR] 3.7 for VPA ≤1500 mg/day versus OR 10.9 for VPA >1500 mg/day) [46], a finding also seen in other studies [49,50,55].
●A case-control study using the European Surveillance of Congenital Anomalies database (registering a combined total of 98,075 major congenital malformations) found that VPA was associated with an increased risk for several congenital malformations [79]. Compared with no ASMs, adjusted ORs for specific malformations were: spina bifida (12.7), atrial septal defect (2.5), cleft palate (5.2), hypospadias (4.8), polydactyly (2.2), and craniosynostosis (6.8). ORs were similarly elevated for the comparison of VPA use with other ASM use.
The effect of VPA on malformation risk is dose dependent, but a lowest safe dose has not been established [56,57,80,81].
Rare cases of neonatal coagulopathy due to VPA-induced hypofibrinogenemia have been reported [82,83].
Phenytoin — In the and EURAP and NAAPR registries, the overall rates of major congenital malformations among pregnancies exposed to phenytoin monotherapy were 2.9 and 6.4 percent, respectively [50,72]. Orofacial clefts, cardiac malformations, and genitourinary defects are the major malformations described with phenytoin [13,48,56,68,76,84,85].
Phenobarbital — Malformations of the heart, orofacial, and urogenital structures occur with increased frequency with phenobarbital [13,48,76,84-87]. A study from the North American Antiepileptic Drug Pregnancy Registry reported a 5.5 percent incidence of major malformations among 199 pregnancies associated with phenobarbital use, a rate that was somewhat higher than for unexposed pregnancies and for those exposed to lamotrigine [72].
Carbamazepine — Carbamazepine is associated with a relatively modest rate of major malformations overall, but the increased rates include neural tube defects [13,74,88]. As much as 0.9 percent of fetuses exposed in utero to carbamazepine have neural tube defects (a sevenfold increase compared with the general population) [48,74,89,90]. However, in the United Kingdom Epilepsy and Pregnancy Registry, compared with other ASMs, carbamazepine was associated with the lowest rate of major malformations: 2.2 percent of 900 pregnancies [47]. The incidence of neural tube defects in carbamazepine-exposed pregnancies was 0.2 percent in this study. Similarly, the rates of fetal malformations associated with carbamazepine monotherapy in the NAAPR, the EUROCAT Antiepileptic Study Database, and EURAP were 2.9, 3.3, and 5.5 percent, respectively [50,52,72,88]. Genitourinary malformations, cardiac malformations and oral clefts have also been reported to be higher than in the general population [76].
Topiramate — Across multiple studies, the estimated absolute risk of oral clefts in a topiramate-exposed pregnancy has ranged from approximately 4 to 29 per 1000 births, compared with an expected unexposed risk in the background population of 1 to 2 per 1000 births [71,72,91-94]. At the lower end of this range, a population-based study nested in the United States Medicaid Analytic eXtract found that the risk of oral clefts at birth was 4.1 per 1000 in the infants born to women exposed to topiramate compared with 1.1 per 1000 in the unexposed group (risk ratio [RR] 2.90, 95% CI 1.56-5.40) [71]. The increased risk of oral clefts associated with use of topiramate early in pregnancy was more pronounced in women with epilepsy, who used higher doses, and is likely another example of a dose-dependent risk for a major malformation.
The risk of fetal malformations is increased further in association with polytherapy regimens that include topiramate. In an Australian registry cohort study, for example, the fetal malformation rate was more than two times higher in polytherapy pregnancies that included topiramate compared with those that did not (14.9 versus 6.6 percent) [51].
Topiramate use in pregnancy is associated with an increased risk for fetal growth restriction and low birth weight [8]. In studies from the North American Antiepileptic Drug Pregnancy Registry, the prevalence of small for gestational age (SGA) was 18 percent in topiramate-exposed neonates, compared with 7 and 5 percent in lamotrigine-exposed neonates and unexposed controls, respectively [22,66]. The risk of SGA was significantly elevated for topiramate (RR 3.5, 95% CI 2.1-5.7) compared with controls after adjusting for multiple factors, including maternal age, parity, smoking status, education, and periconceptional folic acid intake.
Lamotrigine — Data on lamotrigine exposure are mostly reassuring, with fairly good consistency across studies from different parts of the world; three registries reported a 1.9 to 3.2 percent incidence of major malformations after first-trimester exposure to lamotrigine monotherapy [47,52,72,76,92,95-98]. This is similar to estimates from general population-based cohorts.
However, the data regarding risk and dose of lamotrigine have been conflicting. While the International Lamotrigine Pregnancy Registry found no effect of lamotrigine dose, up to 400 mg/day, on the incidence of major malformations [97,99], the United Kingdom Epilepsy and Pregnancy Registry did observe a dose relationship and suggested that doses higher than 200 mg/day might present a risk similar to that of valproate [47]. In the EURAP Epilepsy and Pregnancy Registry, the incidence of fetal malformations with lamotrigine at any dose was low (3 percent), but was significantly higher for patients taking >325 mg per day at conception than for patients taking lower doses (4.3 versus 2.5 percent) [50]. They did not have data on doses during the remainder of the first trimester or lamotrigine blood concentrations.
A report from the NAAPR noted a higher-than-expected prevalence of cleft palate and/or cleft lip in infants exposed to lamotrigine in the first trimester (8.9 per 1000 compared with the expected 0.7 to 2.5 per 1000) [100]. However, population-based case control studies derived from a European congenital anomaly register found that lamotrigine monotherapy did not specifically increase the risk of isolated orofacial clefts relative to other malformations [101,102]. These studies were not designed to study whether lamotrigine was associated with an overall risk of malformations and therefore could not determine an association.
Data released from the manufacturer's pregnancy registry in January 2008 reported that the risk of major malformations after exposure to lamotrigine used in polytherapy with other ASMs, not including VPA, was similar to that of lamotrigine monotherapy but was increased in those exposed to lamotrigine and VPA (16 of 143 pregnancies, 11 percent) [96]. These findings were similar in the final report of the International Lamotrigine Pregnancy registry [97].
Levetiracetam — Levetiracetam use in pregnancy appears to be associated with a low risk of major malformations when used as monotherapy. The largest body of data is from EURAP; the prevalence of major congenital malformations among 599 pregnancies exposed to levetiracetam monotherapy was 17 (2.8 percent) [50]. The United Kingdom and Ireland Epilepsy and Pregnancy Registers reported a risk of only 0.7 percent in 304 pregnancies exposed to levetiracetam monotherapy [103], which is lower than the 2.4 percent risk reported in 450 pregnancies exposed to levetiracetam followed by the North American Antiepileptic Drug Pregnancy Registry [72]. Higher rates have been reported in pregnancies exposed to levetiracetam as part of a polytherapy regimen [103].
Limited data suggest that intrauterine exposure to levetiracetam monotherapy does not adversely affect early developmental outcomes, in contrast with medications such as valproate [104-106]. (See 'Neurodevelopmental risks of ASMs' below.)
Oxcarbazepine — In a registry-based study in Denmark, major congenital anomalies were detected in 11 of 393 infants (2.8 percent) exposed to oxcarbazepine, a rate that was not significantly higher than in non-exposed infants [92]. In the EURAP registry, 10 of 333 infants (3 percent) had a detected major congenital malformation with oxcarbazepine monotherapy in utero exposure [50].
Pregabalin — Animal studies and at least one prospective cohort study have raised concern about possible risk of major congenital anomalies related to first-trimester exposure to pregabalin [107]. In a multicenter observational study, first-trimester exposure to pregabalin (mostly for indications other than seizure disorder) was associated with an increased rate of major congenital anomalies compared with no exposure (6 versus 2 percent) [107]. The small number of exposed pregnancies (n = 164), high rates of polypharmacy and tobacco use among women taking pregabalin, and differences in medical comorbidities between exposed and unexposed women limit confidence in these findings, and more studies are needed [108].
Gabapentin — Several small case-control studies as well as larger registry-based studies that include a limited number of gabapentin-exposed pregnancies have found no association between gabapentin and major congenital anomalies [72,92,109,110]. A prospective cohort study reported on 223 gabapentin-exposed pregnancies compared with 223 unexposed [109]. Rates of major malformations were similar. However, the gabapentin-exposed group had higher rates of preterm births and low birth weight <2500 g. The admission rate to either the neonatal intensive care unit or special care nursery for observation with continued gabapentin use was 38 percent, versus 3 percent in the comparison group (p <0.001). However, only one-third of women were prescribed gabapentin for epilepsy, and most pregnancies involved other neuropsychiatric medications.
Zonisamide — The NAAPR reported that the rate major malformations among 136 pregnancies on zonisamide monotherapy was 1.5 percent [111].
In other reports from the North American AED Pregnancy Registry, the prevalence of small for gestational age (SGA) was 12 percent in zonisamide-exposed neonates, compared with 7 and 5 percent in lamotrigine-exposed neonates and unexposed controls, respectively; (RR 2.2, 95% CI 1.1-4.4) [22,66].
Other medications — There is limited human information on the fetal risks of other ASMs (eg, lacosamide, clobazam, felbamate, tiagabine). While animal and other preliminary data are reassuring [112], patients should be counseled that the actual risks are largely unknown.
NEURODEVELOPMENTAL RISKS OF ASMS — Accumulating evidence from observational studies suggests that antiseizure medication (ASM) treatment during pregnancy leads to deleterious cognitive and behavioral effects that manifest later in the life of the offspring; however, the amount of neurodevelopmental risk varies with the type of ASM and possibly the level of in utero exposure [43,113-120].
Risks associated with ASM use can be minimized by preconception planning and careful management during pregnancy. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period".)
●Valproate – Valproate is the most strongly associated with adverse neurodevelopmental outcomes [8,13,106,121-132]. Several studies illustrate these risks:
•The neurodevelopmental effects of AEDs (NEAD) study, a prospective study of 309 children who had been exposed to ASMs in utero, found that at age three years, children exposed to valproate had intelligence quotient (IQ) scores that were on average six to nine points lower than those exposed to lamotrigine, phenytoin, or carbamazepine [123]. These findings persisted at six years [129]. In addition, the relationship between IQ score and valproate exposure was dose related, and that children's IQ scores were related to maternal IQ scores in all exposure groups except for valproate [124].
•A prospective blinded study of 182 children of mothers with epilepsy and 141 controls found significantly lower verbal intelligence associated with exposure to valproate and ASM polytherapy, but not to carbamazepine monotherapy [125]. In another registry-based study, 102 school-aged children exposed to valproate in utero had lower than mean average language test scores on blinded assessments [126]. Similarly, in a prospective study of 172 infants (mean age of 15 months), in utero exposure to valproate monotherapy was associated with lower mental and motor developmental quotients compared with carbamazepine monotherapy [127].
In utero exposure to valproate also appears to increase the risk of autism spectrum disorders [8,128,133-136]. In a population-based study of more than 650,000 children born in Denmark from 1996 through 2006 that included 6584 women with epilepsy, 2655 ASM-exposed pregnancies, and 508 valproate-exposed pregnancies, the risk of autism spectrum disorder was increased in children exposed to valproate in the overall study population (hazard ratio [HR] 2.9, 95% CI 1.7-4.9; absolute risk 4.4 percent), and there was a trend towards increased risk in valproate-exposed children born to mothers with epilepsy (HR 1.7, 95% CI 0.9-3.2; absolute risk 4.2 percent) [136]. For the more narrowly defined diagnostic code of childhood autism, the relative risk was increased fivefold (absolute risk 2.5 percent). Analyses were adjusted for parental age at conception, parental psychiatric history, gestational age, birth weight, sex of the child, congenital malformations, and parity. An increased risk of autism spectrum disorders was also observed in children exposed to valproate whose mothers had diagnoses other than epilepsy, but not in children of women who were previous users of valproate but stopped at least 30 days before conception; both factors increase confidence in the association. An increased risk was not observed in association with other ASMs, including carbamazepine, oxcarbazepine, lamotrigine, and clonazepam, although the number of events in these subgroups was low.
●Others – A report from the multicenter MONEAD cohort study, a continuation of the NEAD study, evaluated outcomes at two years of age for 292 children of woman with epilepsy and 90 children of healthy women [137]. MONEAD enrolled women from December 2012 to January 2016, and most of the women with epilepsy were using lamotrigine and/or levetiracetam. There were no differences between groups in cognitive outcomes, suggesting that fetal exposure to newer ASMs is associated with a lower neurodevelopmental risk than valproate.
Earlier data also suggested that carbamazepine and lamotrigine are associated with a low or no risk of adverse neurobehavioral or cognitive outcomes [8,138]. The majority of studies support normal cognitive performance in children exposed to carbamazepine in utero [13,118,124,139-141]. Findings from small studies suggest that levetiracetam monotherapy exposure in utero is not associated with later cognitive deficits [105,106].
There are few data regarding in utero exposure to topiramate with regard to its impact on long-term cognitive outcomes [8,138]. Reports from other studies do implicate phenytoin, primidone, and phenobarbital exposures with impaired cognitive performance later in life, although the phenytoin monotherapy-exposed children had normal neuropsychometric results in the NEAD study [13,118,124,139-142]. In the NEAD study, IQ at 4.5 years was lower for children exposed to valproate compared with carbamazepine, phenytoin, or lamotrigine, but verbal abilities were impaired compared with nonverbal skills for all four drugs [124]. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period", section on 'Choice of antiseizure medication'.)
●Polytherapy – ASM polytherapy may present a higher risk of cognitive impairment than monotherapy [13,117,139], and verbal abilities may be particularly affected [124,143].
●Risk with malformations – Some studies find that the risk of cognitive impairments with ASM exposure is highest in children who also have one or more of the major malformations described above [75,144,145].
INHERITANCE OF EPILEPSY — The prevalence of epilepsy is approximately 0.5 to 1 percent of the general population. The likelihood of developing epilepsy is modestly higher in children of a parent with epilepsy, and the degree of risk varies with the type of epilepsy syndrome [146-150]. The ability to stratify risk based on specific genetic findings is just beginning to be clinically available in a very small subset of families with epilepsy. This is likely to change dramatically in the future. Especially during preconception counseling, it is important to consider if either parent could have an epilepsy syndrome linked to a genetic disorder.
The best estimates of familial risk come from a population-based study that included 660 individuals with epilepsy and nearly 2500 first-degree relatives living in a single county in the northern United States [150]. The cumulative incidence of any form of epilepsy by age 40 years was 4.5 percent among individuals with a first-degree relative with epilepsy, which was threefold higher than the incidence in the general population. The risk was highest for relatives of individuals with generalized epilepsy (standardized incidence ratio [SIR] 8.3) and lowest for relatives of those with focal epilepsy (SIR 2.6).
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Seizures and epilepsy in adults".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Epilepsy and pregnancy (The Basics)")
SUMMARY AND RECOMMENDATIONS — While the majority pregnancies in women with epilepsy will be uncomplicated, there is a higher risk of certain complications.
●Morbidity and mortality are increased during pregnancy among women with epilepsy compared with the general population across a range of maternal and neonatal outcomes including preeclampsia, preterm labor, bleeding, placental abruption, poor fetal growth, prematurity, fetal death, and maternal mortality. The magnitude of the increase in risk appears to be relatively small for most complications (ie, between 1 and 1.5 times expected rates), with the exception of maternal mortality, which may be as much as 10-fold higher among women with epilepsy during the delivery hospitalization. Despite the large increase in relative risk of maternal death, the increase in absolute risk is less than 0.1 percent. (See 'Perinatal morbidity and mortality' above.)
●Seizures, particularly convulsive seizures, are believed to be harmful to the fetus. Fetal bradycardia has been documented during maternal seizures. However, data are limited, and more detailed studies are needed of the risk that seizures pose to the fetus. (See 'Effect of seizures on the fetus' above.)
●At least half of individuals with epilepsy will have no alteration of their seizure pattern during pregnancy, especially if nonadherence and sleep deprivation are minimized, and the individual antiseizure medication (ASM) target concentrations are maintained with therapeutic drug monitoring. The main risk factors for increased seizures during pregnancy include baseline seizure frequency before pregnancy and focal epilepsy syndromes. Others include ASM type, polytherapy, and medication adherence. Altered pharmacokinetics of ASMs during pregnancy can contribute to increased seizure frequency if therapeutic drug monitoring is not used to maintain the target blood concentration. (See 'Effect of pregnancy on seizures' above.)
●Major congenital malformations are more common in fetuses exposed to ASMs in utero compared with offspring of untreated women with epilepsy and women without epilepsy. The overall risk of major malformations is 4 to 6 percent, but this varies substantially by the ASM prescribed. Many ASM polytherapy regimens also increase the risk. The timing (early versus late in gestation) is important. (See 'Effect of ASMs on the fetus and neonate' above and 'Major congenital malformations and their risk factors' above.)
●Valproate should be avoided in pregnancy if possible. Across all pregnancy registries, valproate monotherapy is associated with the highest rates of major malformations. Phenytoin and phenobarbital have also been associated with relatively high baseline rates of major malformations. The spectrum of structural and neurodevelopmental teratogenic risks for different ASM monotherapies is depicted in the figure (figure 3). There are not enough data on the other ASMs to place them along the spectrum. (See 'Risks with specific ASMs' above.)
●In utero exposure to some ASMs is associated with impaired cognitive and behavioral development. The evidence for this association is strongest with valproate, which has been associated with impaired cognitive development as well as an increased risk of autism and autism spectrum disorders. (See 'Neurodevelopmental risks of ASMs' above.)
●Risks associated with pregnancy and epilepsy can be minimized by preconception planning and careful management during pregnancy. (See "Management of epilepsy during preconception, pregnancy, and the postpartum period".)
36 : Effects of monitoring strategies on seizures in pregnant women on lamotrigine: a meta-analysis.
59 : Recurrence risk of congenital malformations in infants exposed to antiepileptic drugs in utero.
