Anticonvulsants

Compared to lithium, higher teratogenic risk has been reported to be associated with anticonvulsants. The risk for major birth defects in infants born to women receiving anticonvulsants has been reported to be twice that seen in general population.[47] Further it is seen that the increased risk for teratogenesis is associated with high maternal serum anticonvulsant levels and exposure to multiple anticonvulsants at a time.[48,49]

Valproate

Valproate is now considered as a mood stabilizer. It is effective in management of acute manic and depressive episodes and is also useful in prevention of relapse of both manic and depressive episodes. Valproic acid and its various derivatives have been known to have an increased risk of causing neural tube defects in the range of 1.0–5.0%, which is about 2–10-fold higher than the general population base rates of about 0.5%.[50,51] These risks are of concern because neural tube formation occurs within the 1^st month of gestation, a time period during which the pregnancy is not even diagnosed. The neural tube defect with valproate more often involves the lumbosacral rather than the anencephalic region, suggesting the effect of valproate on the closure of neural crest.[52] Prenatal exposure to valproate has been associated cardiovascular malformations, intrauterine growth retardation, genital anomalies, hydrocephalus, limb defects (radial ray effects, fibrous aplasia of lower limbs), and pulmonary atresia.[47,48,49,50,51,52,53,54,55] In a prospective study, Wyszynski et al.,[53] evaluated the rate of occurrence of major malformations at birth in infants of mothers who had taken valproate as monotherapy and had enrolled in the North American Antiepileptic Drug (AED) Pregnancy Registry. Of the 149 valproate monotherapy exposed pregnancies, major malformations were seen in 16 cases. When these data were compared with the major malformations with other anticonvulsants, it was concluded that valproate was associated with more frequent adverse effects. Another recent case–control study, which was based on European Surveillance of Congenital Anomalies antiepileptic-study database, evaluated the outcome of exposure to valproic acid monotherapy in 180 registrations. Compared to no use of antiepileptic medications, use of valproic acid during the first trimester was associated with increased risk of spina bifida, atrial septal defect, cleft palate, hypospadias, polydactyl and craniosynostosis.[56] Emerging data also suggest that the risk of malformations with valproate is dose related[57,58] with higher risk associated with use of valproate in excess of 1000 mg/day.[57] Data based on the review of United Kingdom (UK) and Ireland Epilepsy and Pregnancy Registers have estimated the risk of major congenital malformations with valproate monotherapy to be 6.7%, which is higher than that reported for carbamazepine (2.6%) and lamotrigine (2.3%).[59]

Use of valproate close to delivery is associated with withdrawal symptoms of abnormal tone, feeding difficulties, irritability and jitteriness.[54] In addition, neonatal complications like heart rate decelerations,[60] liver toxicity,[61] hypoglycemia,[62] and reductions in neonatal fibrinogen levels[63] have also been reported. Long-term behavioral and cognitive outcome with exposure to valproate in pregnancy has been evaluated. Meador et al.[64] evaluated the IQ scores at the age of 3 years of 258 children who were exposed to anticonvulsants during pregnancy. IQs were lower for those exposed to valproate compared to carbamazepine, LTG, and phenytoin, after controlling for maternal IQs and blood levels of anticonvulsant medications. It was further found that maternal valproate blood level had significant inverse correlation with cognitive functioning. A follow-up of the children at 4.5 years showed the persistence of lower IQ, which was negatively associated with dose of valproate[65] Subsequently, a large scale study evaluated IQ of children at the age of 6 years and provided credence to this finding.[66]

A study also suggested that compared to the children exposed to carbamazepine and lamotrigine, children exposed to valproate prenatally have lower cognitive fluency and originality.[67] Recent studies have also reported neurodevelopmental delay in children exposed to valproate prenatally.[68,69,70] Another cause of concern is high rate of autism spectrum disorder among the infants prenatally exposed to valproate monotherapy with absolute risk of 4.42% for autism spectrum disorder and 2.5% for childhood autism.[71,72] A recent review of data also suggests increase in the risk of attention deficit hyperkinetic disorder with antenatal exposure of valproate.[73]

With regard to secretion of valproate in breast milk, it has been shown that valproate is minimally present in breast milk. Piontek et al.[74] evaluated the level of valproate in 6 breast-fed mother-infant pairs and reported that serum valproate levels in the infants ranged from 0.9% to 2.3% of the mother's serum level. In another study of 2 infants whose breastfeeding mothers were taking valproate, Wisner and Perel[75] reported that valproate concentrations in the infants were 1.5% and 6%, respectively. The American Academy of Neurology and American Academy of Pediatrics support breastfeeding if the mother is taking valproate.[76,77]

Carbamazepine

Use of carbamazepine during the first trimester of pregnancy has been shown to be associated with the risk of neural tube defects at a rate of about 0.5–1.0%.[78,79] Infants exposed to carbamazepine prenatally are also at increased risk for craniofacial abnormalities, fingernail hypoplasia, developmental delay, growth retardation, microcephaly, spina bifida, and cardiac abnormalities.[47,80] Recent data suggest that the risk increases in a dose–dependent pattern with higher risk with doses 400 mg/day.[81] Transient hepatic toxicity (cholestatic hepatitis and direct hyperbilirubinemia) in neonates exposed to carbamazepine during pregnancy has also been noted.[80,82] In terms of behavioral and cognitive outcome, studies have shown that carbamazepine exposure during pregnancy and neonatal period does not lead to significant cognitive dysfunction in childhood.[83,84]

The data on serum concentrations of carbamazepine among breastfeeding infants are largely based on assessment of newborns of mothers who were on carbamazepine during the pregnancy. Evidence suggests that in general the concentration of carbamazepine in breast milk is much lower than the maternal serum levels, with concentrations in infants ranging from 6% to 65% of maternal levels.[85] Studies that have evaluated carbamazepine exposure in nursing infants have reported occasional cases of transient hepatic dysfunction.[80,82] Both American Academy of Neurology and American Academy of Pediatrics support breastfeeding if the mother is taking carbamazepine.[76,77]

Oxcarbazepine

Oxcarbazepine, a congener of carbamazepine has been used for the treatment of bipolar disorders. In a review of data, Eisenschenk[86] reported that, in the 248 patients identified in the literature as receiving oxcarbazepine during pregnancy, 6 malformations were reported; indicating a malformation risk of 2.4%, which is within the 2–4% malformation rate seen in the general population. It is noteworthy that in 4, out of the 6 pregnancies with congenital malformation, the mothers were also receiving adjunctive antiepileptics. The malformations reported include major cardiac malformation (adjunctive therapy with phenobarbital), ventricular septal defects (1 monotherapy and 1 adjunctive therapy with lamotrigine), facial malformations (cleft soft palate, facial dysmorphism, terminal transverse defect, disproportion of head and trunk, and extrophia vesicae) in 1 case, congenital hydronephrosis (adjunctive therapy with vigabatrin), major urinary tract defect (1 case) and 1 case of spina bifida cystica and clubfoot (adjunctive therapy with valproate and clobazam).

Lamotrigine

In recent time use of lamotrigine in bipolar disorders has increased. However, most of the data about its safety in pregnancy is available from its use in pregnant females with epilepsy. Newport et al.[87] pooled the data from the available studies and reported 2.6% risk (78/2974) of major fetal anomalies following first trimester exposure to lamotrigine, which is within the range of anomalies among births reported in the general population. The obstetrical outcome information is maintained by the International lamotrigine Pregnancy Registry. Cunnington et al.[88] reported that of the 414 first trimester exposures to lamotrigine monotherapy, about 3% (12 offspring) had major birth defects which are similar to that in general population. However, no distinctive pattern of major birth defects was apparent among the birth defects detected following exposures to lamotrigine monotherapy. They found a higher rate of major birth defects (12.5%) in 88 first trimester exposures to combination of lamotrigine and valproate. Recently, some data have emerged to suggest that exposure to LTG is associated with increase in the risk of oral clefts. Holmes et al.[89] presented the data from North American AED Pregnancy Registry with regard to lamotrigine. Of the 684 infants exposed to lamotrigine, 16 (2.3%) had major malformations. Five infants (7.3/1000) had oral clefts (isolated cleft palate [3 cases], isolated cleft lip [1 case], and cleft lip and palate [1 case]). The rate among the lamotrigine-exposed infants was 10.4-fold higher in comparison to unexposed infants (n = 206,224) surveyed at birth at Brigham and Women's Hospital in Boston. However, another study based on data from UK Epilepsy and Pregnancy Register, did not confirm this finding. The authors reported data of 1229 pregnancies exposed to lamotrigine as monotherapy, resulting in 1151 live births. Twenty-eight major congenital malformations were reported with no specific pattern; of which only one male infant was born with nonsyndromic cleft lip and palate.[90] Recent data suggest that the risk increases in a dose dependent pattern with higher risk with doses 300 mg/day.[81] Studies in women with bipolar disorder treated with lamotrigine during pregnancy suggest lower serum level to dose ratios during pregnancy than during the postpartum period.[91] Studies also suggest higher clearance level of lamotrigine during pregnancy and suggest the need for dose adjustment.[92] In terms of neurobehavioral effects a follow-up study of 23 infants, reported no neurobehavioral effects with lamotrigine at 12 months of age.[93] However, one recent report suggested behavioral outcomes comparable to valproate. The children had reduced nonverbal IQ scores, along with lower scores on motor and sensory measures, lower scores on executive functions, behavioral measures and attentional measures as per the parent-report.[94]

Few case reports and case series reported mean milk/plasma ratios ranging from 0.40 to 0.61 for lamotrigine.[95,96,97,98,99] One recently published study involving 30 lactating mothers and their infants reported a mean milk/plasma ratio of 41.3% with a range of 5.7–147%. The authors also reported higher lamotrigine concentrations in breast milk 4 h after maternal dose, although this finding was not statistically significant. Compared with maternal plasma concentrations, the plasma concentrations in infants were 18.3%. Based on these findings authors reported infant lamotrigine dose to be 0.51 mg/kg/day, and compared to maternal dose and the relative infant lamotrigine dose was 9.2%. Almost all infants (7 out of 8) were found to have mild thrombocytosis at the time of serum sampling. No other adverse events were observed or reported in the breast-fed infants.[100]

Topiramate

Animal studies have demonstrated craniofacial and skeletal abnormalities.[101,102] Kwarta et al.[103] reported 2 malformations (micrognathia, phimosis) in 29 pregnancies exposed to topiramate monotherapy. Hunt et al.[104] reported pregnancy outcome in patients receiving topiramate based on UK Epilepsy and Pregnancy Register. In the 70 pregnancies exposed to topiramate monotherapy, 3 major malformations were seen; 2 infants had cleft lip and cleft palate and 1 had hypospadias. It was also seen that women who gave birth to neonates with major malformation were receiving relatively higher doses (mean - 400 mg vs. 238 mg) compared to women who gave birth to children without malformation. It was also seen that out of 61 cases exposed to monotherapy and for whom data was available, 6 infants were born at 37 weeks of gestation or less. Further, of the 56 cases exposed to monotherapy and for whom full data on gestational age and birth weight was available, 8 (14.3%) were small for gestational age. However, in women receiving topiramate as part of polypharmacy, 13 major malformations were reported. It was observed that major malformation rates were higher when topiramate was given along with valproate than with other antiepileptics. A large sample study based on the Global Medical Safety database, reported pregnancy outcome of 589 cases. The most common outcome was a live birth, noted in 75.55% of pregnancies. However, this outcome varied based on the diagnosis for which topiramate was used, with highest proportion of live births seen in patients with epilepsy.[105] The risk of major malformation with topiramate has been reported to be 4.2–9%.[105,106] Among the major malformations reported cleft lip or palate anomalies, limb, hand, or other skeletal anomalies and respiratory or cardiovascular anomalies were more frequently documented.[105] Other reports also suggest association of first trimester use of topiramate and oral cleft lip/palate anomalies.[107] Other abnormalities related with topiramate use include small for gestational age[108,109,110] and microcephaly.[109]

Gabapentin

Animal studies have demonstrated delayed bone ossification, hydronephrosis and increased rates of hydroureter.[101] No information is available regarding its possible teratogenic effects in humans. Ohman et al.[111] evaluated the outcome of 6 pregnancies in women receiving gabapentin during pregnancy and reported uneventful deliveries with birth to healthy children in 5 out of the 6 pregnancies. In 1 pregnancy, there was a premature delivery at 33 weeks of gestation. In terms of postdelivery complications, mild hypotonia and cyanosis was seen in 1 infant, which started 8 h after delivery. However, the infant was discharged from hospital after 4 days in a completely normal state. One recent study compared the pregnancy related outcome of 223 gabapentin exposed pregnancies with unexposed pregnancies and reported higher rate of preterm birth and low birth weight with gabapentin. Further those neonates who were exposed to gabapentin near the delivery more frequently required neonatal intensive care treatment. Two neonates exposed to gabapentin were also reported to have suffered from poor neonatal adaptation.[112]

With regard to transplacental transfer of gabapentin Ohman et al.[111] reported accumulation of gabapentin with umbilical cord/maternal plasma concentration ratios ranging from 1.3 to 2.1 (mean - 1.7). However, the gabapentin plasma concentrations declined with an estimated half-life of 14 h in the newborns. Studies which have estimated the concentration of gabapentin in milk suggest that mean milk/maternal plasma concentration ratio is 1.0 (range, 0.7–1.3) from 2 weeks to 3 months. Accordingly, the infant dose is estimated to be 0.2–1.3 mg/kg/day, which is about 1.3–3.8% of the dose received by the mother. When the plasma concentrations in the breast-fed infants were estimated, it was reported that the concentration in the infant was about 12% of the mother's plasma levels. However, this concentration was not associated with any adverse effects.

Atypical antipsychotics

Atypical antipsychotics are quite frequently evaluated for the treatment of various phases of mood disorders. However, there is relatively little data available with regard to the safety of these agents in pregnancy and lactation. In this section, the data with regard to use of atypical antipsychotics in pregnancy and lactation is reviewed, irrespective of the disorder in which these were evaluated.

Olanzapine

Animal studies done in rodents did not reveal any evidence of teratogenicity with olanzapine despite the use of doses higher than those used clinically. There are few studies which have evaluated the use of olanzapine during pregnancy. Goldstein et al.[113] reported outcomes of 37 olanzapine-exposed pregnancies ascertained prospectively. Of the 37 pregnancies, 14 were terminated by therapeutic abortions with no abnormalities reported in the fetus. In the rest of the pregnancies (n = 23), normal birth without complications occurred in 16 cases; spontaneous abortion occurred in 3; stillbirth in 1; prematurity in 1; postmaturity in 2 cases, with one of new born developing perinatal complications in the form of meconium aspiration after cesarean section; and no major malformation in any case. The rates of complications were less than or comparable to the range of base rates for general population. Additionally, the authors also reported 11 retrospectively identified cases, in which major malformations in the form of dysplastic kidney (1 case) and Down syndrome (1 case) was found. There were 2 spontaneous abortions and 1 fetal death. Perinatal complications were noted in 3 cases and postperinatal complications in 2 cases. Overall, out of 11 cases, no complications were observed in only 1 case. In a review, Gentile[114] reviewed the data of 248 reported cases on olanzapine maintained by the manufacturer (Eli Lilly) up to December 2006. Among these, spontaneous abortion was reported in 24 cases and perinatal complications in 49 cases. Various malformations which had been recorded include renal malformation (n = 5), additional thumb digit (n = 2) and one case each of bilateral talipes, spontaneous abortion of severe deformed fetus, pretragus fibrochondroma, clubfoot, anencephaly, absent heart, cleft palate, ventricular septum defect, albino infant, esophageal atresia, myelomeningocele plus hydrocephalus, and absent fingers. However, it was also noted that some of the cases in which malformations were found had also received other concomitant medications. Further, the authors pointed out that the manufacturer concluded that the prevalence of adverse pregnancy outcome did not differ from that found in the general population. In a recent review of data maintained by Eli Lilly, outcome of data of 610 pregnancies exposed to olanzapine was presented. The authors reported normal birth in two-third (65.7%) of cases, premature birth in 9.8%, spontaneous abortion in 9.3% and congenital anomalies in 4.4% of pregnancies and perinatal complications were seen in 8% of pregnancies.[115]

Levinson et al.[116] found 8 live births with no malformations, 1 spontaneous abortion, and one stillbirth. In a study which evaluated the outcome of 79 olanzapine exposed pregnancies, major malformations in the form of craniosynostosis and ureteral reflux (n = 1), hand/finger reduction (n = 1), ventricular septum defect and upper alimentary tract malformation (n = 1) were observed.[117] In another study involving 60 pregnancies exposed to olanzapine, 1 case of major malformation (midline defects, cleft lip, encephalocele, and aqueductal stenosis) was reported.[118]

Various other case reports and case series have also reported birth of healthy infants without complications despite prenatal exposure to olanzapine.[119,120,121,122,123,124,125,126,127,128,129,130,131,132] A few case reports have reported defects in the form of atrioventricular canal defect and unilateral clubfoot,[133] hip dysplasia,[134] meningocele and ankyloblepharon[135] and microcephaly and anophthalmos.[136] In some of these studies, case series and case reports development or worsening of gestational diabetes mellitus was also noted.[125,126,127] Perinatal syndrome reported with use of olanzapine in pregnancy includes preterm delivery, low birth weight, hypotonia, neonatal respiratory complications, neonatal cardiovascular complications,[129] Erb's palsy, jaundice[128] and high birth weight/large for gestational age.[128,129] With respect to the postnatal long-term effects, Gati et al.[137] did not find any postnatal behavioral sequelae in children up to the age of 6–10 years.

Few studies have evaluated the excretion of olanzapine in breast milk. The relative dose in infant has been reported to be in the range of 0.66–1.19%.[138,139] Data from about 40 cases are available with respect to olanzapine use during breast feeding. Goldstein et al.[140] reported 26 cases in which breast feeding was continued for 1–8 months. Four infants developed adverse reactions in the form of jaundice and sedation (1 case), shaking, poor sucking and lethargy (1 case), protruding tongue (1 case) and diaper rash, diarrhea, and sleep disorder (1 case). Kirchheiner et al.[120] reported temporary motor delay in a case whose mother took olanzapine 10 mg/day during the lactation. However, other case reports and case series have not reported any untoward effects on the infant with continuation of olanzapine in the dose range of 2.5–20 mg/day during breast feeding.[128,138,141,142] In the recently published study, data were available for 102 newborns/infants exposed to olanzapine through breast milk. Adverse outcome in the neonate or infant were reported in 15.6%, with commonly reported adverse events in the form of sedation (3.9%), irritability (2%), tremors (2%) and insomnia in 2% of neonates/infants.[115] Studies that have followed up children exposed to olanzapine through breast milk at 1–2 years of age, report no increase in adverse outcomes when compared to control group.[143]

Risperidone

Animal studies in rats have shown increased incidence of pup deaths and stillbirth with use of risperidone during pregnancy.[101] Rosengarten and Quartermain[144] reported impaired learning and disrupted short-term retention in adulthood with use of risperidone in dam during pregnancy. In their study of 49 risperidone exposed pregnancies in humans McKenna et al.[118] did not report a single case of major malformations. Similarly, many studies have reported no major malformations with use of risperidone.[145,146,147,148,149,150] However, there are a few case reports of agenesis of corpus callosum,[101] and spontaneous abortions.[116] Reis and Källén[117] reported 1 case of major malformation (anal atresia along with pulmonary malformation) in 51 risperidone exposed pregnancies. The postmarketing surveillance data of 713 pregnancies (prospective data for 516 and retrospective data for 197 pregnancies) exposed to risperidone was reported. Majority of the adverse pregnancy and fetal/neonatal outcomes were reported for pregnancies for which retrospective data was available. Among the prospectively reported, out of the 68 pregnancies with known outcome, organ malformations were observed in 3.8% and spontaneous abortions occurred in 16.9% pregnancies. When compared to the general population, the rates were comparable.[151] Of the retrospectively reported outcome major organ malformations were reported in 12 pregnancies. Major malformations reported included cleft lip/palate (n = 2), and one case each of esophageal atresia along with ear pinna hypoplasia and slight facial dysmorphia, Ivemark syndrome, Moyamoya disease, ventricular cyst in the brain, patent foramen ovale, hypoplastic left heart, dilated cardiomyopathy, right auricular achondroplasia, mild talipes equinovarus, gastroschisis, and Pierre-Robin syndrome. However, it is important to remember that in some cases in which major malformations were seen, the mothers also received other concomitant medications.[151] In terms of neonatal outcome, McKenna et al.[118] concluded that there was no difference in the prevalence rate of poor pregnancy outcome and perinatal complications between the risperidone exposed group and a control group exposed to nonteratogens. In the postmarketing surveillance study,[151] perinatal syndrome (in the form of tremor, jitteriness, irritability, feeding problems) was reported in 37 pregnancies, of which 21 cases involved behavioral or motor disorders. Other adverse pregnancy outcomes reported with use of risperidone include oligohydraminos,[150] small for date,[152] and gestational diabetes mellitus.[117] A case report of neuroleptic malignant syndrome in a pregnant women receiving risperidone has also been documented.[153]

With regard to use of risperidone during lactation, studies suggest that relatively low infant dose (0.84–4.7%) of the weight adjusted maternal dose[154,155] and an additional 3.46% of its metabolite are transferred to infant.[155] The data does not suggest any untoward neurodevelopmental problems with risperidone.[150,155,156]

Quetiapine

Delay in skeletal ossifications, reduced fetal weight, and increased fetal and pup deaths have been reported with use of quetiapine in pregnancy in animal models.[101] In human studies, quetiapine has been shown to have the least placental passage (23.8%) compared to both first generation antipsychotics (haloperidol) and other second generation antipsychotics (risperidone and olanzapine). Most of the human studies have not reported any major malformation or neonatal problems with use of quetiapine in pregnancy.[157,158,159,160,161,162,163,164,165,166,167,168] McKenna et al.,[118] reported on the data maintained by the manufacturer till March 2005. A total of 446 pregnancies were exposed to quetiapine. In terms of congenital malformation, out of the 151 pregnancies for which outcome was reported, 8 pregnancies had congenital malformations; with no commonality between the malformations seen. It is important to note that concurrent medications were used in 7 of the 8 cases with congenital anomalies. In a postmarketing surveillance study, out of the 6 cases, spontaneous abortion was reported in 2 cases.[157] McKenna et al.[118] also reported the outcome of 36 pregnancies exposed to quetiapine. Although the authors did not report any specific congenital anomalies associated with quetiapine, they concluded that compared to a control group of women exposed to nonteratogenic agents, quetiapine was not associated with an increase of teratogenic risk.

With regard to excretion of quetiapine in breast milk, studies suggest relatively low dose (0.09–0.43%) of the weight adjusted maternal dose in infant[161,169] and the quetiapine concentration in infant's plasma is estimated to be about 6% of the maternal drug concentration.[169]

Amisulpiride

Data from animal studies on mice and rabbits show lack of teratogenic effect of amisulpride with doses up to 4 times the maximum recommended human dose.[114] However, data on the effect of amisulpride on human pregnancies are lacking.

Ziprasidone

Use of ziprasidone in animals suggests its association with developmental delays, possible teratogenic effects (cardiac, renal, and skeletal), and increased still births.[101] One case report presented the association of ziprasidone with cleft palate,[170] whereas another case report reported on adverse outcome of use of ziprasidone along with citalopram during pregnancy and lactation.[171]

Aripiprazole

In animal studies aripiprazole has shown developmental toxicity, including possible teratogenic effects in rats and rabbits. The main effects were delayed skeletal ossification and decreased fetal weight with 3–10 times the maximum recommended human dose.[172] Only 13 case reports and 1 small case series (3 cases) have discussed the outcome of aripiprazole exposed human pregnancies.[173,174,175,176,177,178,179,180,181,182,183,184] In all cases except one, no teratogenic effects were reported. However it is important to note that in 9 out of the 16 pregnancies aripiprazole was started after first trimester of gestation. In one case minimal hip dysplasia was documented.[184] In terms of neonatal and obstetrical outcome, 1 case required caesarean section due to development of unexplained fetal tachycardia.[174]

Sparse data are available with regard to use of aripiprazole during lactation. Animal studies suggest that aripiprazole is excreted in milk. One case report suggests failure of lactation in women treated with aripiprazole in pregnancy.[173] Another case report suggested lack of aripiprazole secretion in breast milk.[178]

Clozapine

Clozapine is the only antipsychotic which belongs to category B of FDA classification. Animal studies in rats and rabbits reveal no harm to the fetus with doses of approximately 2–4 times higher than the human dose (clozapine prescribing information).[114] Data with regard to use of clozapine in human pregnancy is available only in the form of case reports/series and occasional review. Waldman and Safferman[185] reported at least 15 normal births following maternal exposure to clozapine in pregnancy. Many others also suggest no definitive association between maternal exposure and congenital anomalies in humans.[186,187,188,189,190,191,192] A study which reported outcome of 61 pregnancies exposed to clozapine revealed 5 congenital malformations and 5 perinatal syndromes.[193] Various other associations reported with maternal exposure are decreased variability of fetal heart rate,[194,195] delayed peristalsis,[196] delayed speech acquisition,[197] floppy infant syndrome,[188] gastroesopahgeal reflux disease,[198] intrauterine growth retardation with oligohydramnios and intrauterine death,[199] neonatal seizures[189] and new onset or worsening of gestational diabetes mellitus with shoulder dystocia.[185,187] However, many of these findings have been complicated by the concomitant use of other drugs, malnutrition or family history of diabetes mellitus. Clozapine overdose during pregnancy has also been shown to lead to fatal poisoning in the newborn[200] and to be associated with absent fetal heart rate variability and delayed peristalsis in the newborn.[201]

As with other antipsychotics, data with regard to use of clozapine during lactation is sparse. In one case series, 2 out of the 4 infants breast-fed by mothers receiving clozapine, developed adverse events. In a case report, delayed speech acquisition was reported in the new born whose mother received clozapine during pregnancy and lactation.

Benzodiazepines and pregnancy

Benzodiazepines are used commonly in subjects with bipolar disorders as adjunctive medications for mood stabilization, agitation, and sleep problems. All benzodiazepine compounds diffuse readily across the placenta to the fetus and are secreted in breast milk. The amount excreted depends on the oral bioavailability, plasma protein binding, maternal blood concentrations, ionization, molecular weight, half-life, degree of lipophilicity and pharmacokinetics of each benzodiazepine.[202] The use of benzodiazepines in pregnancy has been reviewed in detail by Iqbal et al.[202] The most important issues associated with use of benzodiazepines in pregnancy shall be discussed here.

The risk of malformation is highest if the fetal exposure occurs between 2 and 8 weeks of pregnancy. Further if benzodiazepines are given at or near term, they may cause fetal dependence and eventual withdrawal symptoms.[202]

Diazepam

In a review of 599 oral clefts it was shown that use of anxiolytics, mostly diazepam, during the first trimester of pregnancy was associated with development of oral clefts.[203] However, this has not been a consistent finding in the later studies, with some reporting similar association,[204,205,206] while others not finding the same.[207,208,209,210] There are case reports of other malformations, however, some of these have been inconclusive because of use of other medications along with diazepam.[202] Use of high doses of diazepam (usually more than 30–40 mg/day) during labor has been associated with withdrawal symptoms starting within a few days to 3 weeks and lasting up to several months in neonates. The withdrawal symptoms are characterized by hypertonia, hyperreflexia, restlessness, irritability, abnormal sleep patterns, inconsolable crying, tremors or jerking of the extremities, bradycardia, cyanosis, suckling difficulties, apnea, risk of aspiration of feeds, diarrhea and vomiting, and growth retardation. Other reports suggest that use of high doses may also be associated with “floppy infant syndrome” characterized by withdrawal symptoms, hypothermia, lethargy, respiratory problems, and feeding difficulties.[202] Use of diazepam during late pregnancy and labor has also been associated with development of prolonged hyperbilirubinemia of the newborn and potentially to kernicterus. This has been linked to the preservative (sodium benzoate) in the diazepam formulation, which competes with diazepam for plasma protein binding.[202]

Although occasional reports have associated the therapeutic use of diazepam with congenital malformation, the bulk of the evidence indicates that the use of diazepam during gestation has no adverse effects on the child's development if used in low doses.[202] However, it is to be remembered that when it is used near term, the dosage should be tapered off over weeks to avoid development of neonatal withdrawal and floppy infant syndrome.

Low dose use of diazepam either just before the delivery or during postpartum by the mother is associated with sedation in the new born. In general, low concentration of diazepam has been found in breast milk; however, the concentration varies with maternal dose. Use of higher doses of diazepam during breast feeding is associated with lethargy and weight loss in the neonate, which improves after stopping breast feeding. Hence, while using diazepam during pregnancy or labor the treating doctors should be aware of the side effects and must take appropriate measures.[202]

Clonazepam

Use of clonazepam in pregnancy has been shown to be associated with apnea, bilateral inguinal hernia, congenital heart disease, cyanosis, hip dislocation, hypotonia lethargy, paralytic ileus of the small bowel, undescended testicle, uteropelvic junction obstruction, and ventral septal defect.[202] However, it is important to note that in most cases clonazepam was used along with other antiepileptics like phenytoin and barbiturates.

Use of clonazepam during lactation has also shown to be associated with apnea, CNS depression, cyanosis, excessive periodic breathing and hypotonia. Hence, clonazepam should be used during pregnancy and lactation only when the clinical benefit to the mother justifies the risks to the fetus and newborn. If used, the fetus and the newborn should be closely monitored.[202]

Lorazepam

Use of lorazepam during pregnancy has been associated with anal atresia and neonatal withdrawal symptoms. The latter can be severe, because of the short half-life of lorazepam, and are characterized by low Apgar scores, depressed respiration, hypothermia, poor suckling and jaundice. Thus, injectable lorazepam should be avoided during pregnancy as far as possible.[202] Oral lorazepam should be used during pregnancy only in life-threatening situations or in cases of severe disease for which other safer drugs cannot be used or are ineffective.

In breast milk, lorazepam is secreted in low quantity and is considered to be relatively safe.[202]

Unplanned pregnancy

Unfortunately, about 50% of pregnancies in women with bipolar disorders are unplanned. In many instances, pregnancy is detected during or after the highest risk period for some agents has passed. Discontinuation of the medication after the risk period has passed may place the woman at risk without conferring any appreciable benefit. The variables which must be taken into account while finalizing the treatment plan include the clinical stability of the patient, duration of pregnancy, psychotropic agent which the patient is taking and treatment preferences of the patient.

If the patient has not completed the first trimester and reports with confirmed pregnancy, the pros and cons about the medication continuation and abrupt stoppage of medication have to be discussed. If the patient is on anticonvulsants, then a high dose of folic acid should be prescribed for the woman. If it is not possible to stop the medication then attempt should be made to reduce the dose to minimum.

If the patient has completed the first trimester of pregnancy, then the patient and the spouse should be explained about the pros and cons of continuation of pregnancy, depending on the agent, which the patient was taking. If the patient had been on polypharmacy, then the additional risk should be emphasized.

If the decision to continue the pregnancy is made, the decision about the mood stabilizer should take past history of response and the risk associated with a particular agent during the second and third trimester. Continuation of the same mood stabilizer with informed consent of the patient and spouse should be preferred. However, some authors suggest that if the woman is on valproate and if there is no history of nonresponse to lithium, the switch to lithium should be considered.[212]

Monitoring while continuation of mood stabilizers during pregnancy

If mood stabilizers are continued during pregnancy, prenatal screening for congenital malformation with a high resolution ultrasound and fetal echocardiography is recommended around 16–18 weeks of gestation to screen for cardiac anomalies. The possibility of fetal neural tube defects should be evaluated with maternal serum alpha fetoprotein and ultrasonography. In addition, use of folic acid is recommended prior to conception and during the first trimester for women receiving anticonvulsants, although it is unknown whether supplemental folic acid can reduce the risk of neural tube defects in the setting of anticonvulsant exposure.[49,213] For patients who are continued on carbamazepine or oxcarbazepine, Vitamin K supplementation (10 mg/day orally) during the last month of pregnancy is recommended. The new born should also receive 1 mg of Vitamin K intravenously or intramuscularly on day 1 after delivery.[76]

Monitoring while continuation of mood stabilizers near the term/labor

Some experts suggest that the dose of lithium should be reduced at the onset of labor to avoid toxicity resulting from the rapid reduction of vascular volume at delivery.[214] Close monitoring of symptoms and serum lithium levels are required to avoid relapse or toxicity during delivery and the immediate postpartum period.[13,29]

Treatment during postpartum

Puerperium is the period of heightened risk of relapse for subjects with bipolar disorders. Hence, restarting of mood stabilizer during immediate postpartum should be considered strongly in subjects with high risk of relapse, as data show that use of lithium as a prophylactic agent in the postpartum period reduces the rate of relapse from nearly 50% to <10%.[215,216]

First episode during pregnancy and puerperium

For women who develop first episode of mania during pregnancy, the psychiatrist is called upon to decide about which medication to start. In such a scenario, the decision about the mood stabilizer is a tricky one and should take into consideration the severity of symptoms, risk to the mother and the fetus. If the patient has not completed first trimester of pregnancy, typical high potency antipsychotic or atypical antipsychotic like clozapine, olanzapine or risperidone should be tried before considering lithium, valproate or carbamazepine. If the women has completed first trimester, then also antipsychotic should be preferred over lithium (because of risk associated during labor), valproate (long-term cognitive side effects) and carbamazepine. If one of these 3 agents has to be used during the second or third trimester, lithium should be preferred over carbamazepine and valproate. The patient and the spouse should be explained about the risk with such medications. If one has to use benzodiazepines, they should be used for shortest possible duration in minimal dose and should preferably be tapered off slowly with monitoring of fetus.