An increased frequency of cleft palate, eye and ear defects, and fetal loss among the offspring exposed to lithium carbonate in utero in utero has been observed in animal teratology studies (Smithberg and Dixit, 1982; Szabo, 1970; Wright has been observed in animal teratology studies (Smithberg and Dixit, 1982; Szabo, 1970; Wright et al et al., 1971). Inconsistencies in animal teratology studies of lithium make it impossible to interpret these data for use in evaluation of human exposures.

SEDATIVES, HYPNOTICS, AND TRANQUILIZERS.

Barbiturates Barbiturates are a family of drugs that are all salts of barbituric acid. To varying degrees, these drugs have a.n.a.lgesic, sedative, and hypnotic actions (Box 10.3). In addition, they have anticonvulsant action.

PHEn.o.bARBITAL.

Phen.o.barbital is a barbiturate used to treat seizure disorders. In the past this drug was used for mild anxiety or sedation, but it is now rarely used for that purpose today.

Sedatives, hypnotics, and tranquilizers 195.

Administration of phen.o.barbital is usually via the oral route, but it may be given parenterally if necessary.

Possible teratogenic effects of phen.o.barbital and phenytoin were suspected early (Janz and Fuchs, 1964). The risk for the pregnant woman treated with phen.o.barbital and other seizure medications of having an infant with congenital malformations is two to three times greater than that of the general population. It is not clear whether the increased risk is secondary to the anticonvulsants, genetic factors, the seizure disorder itself, or possibly a combination of these factors (Kelly, 1984).

Evidence implicates anticonvulsants as the etiology (Hanson and Buehler, 1982). An increased frequency of minor and major congenital anomalies was found among offspring of pregnant women who received phen.o.barbital during gestation for seizure disorders compared to women who received the drug for other reasons (Hanson and Buehler, 1982). The frequency of congenital anomalies was not increased in several studies of children born to women who were treated with phen.o.barbital for epilepsy when compared to the offspring of women with epilepsy who were not treated (Greenberg et al et al., 1977; Nakane et al et al., 1980; Robert et al et al., 1986; Rothman et al et al., 1979). In an a.n.a.lysis of phen.o.barbital monotherapy exposure separately (i.e., no other concomitant anticonvulsants), the frequency of congenital anomalies was not always increased (Nakane et al et al., 1980). In a multinational European collaborative study of 250 infants born to women with epilepsy, the frequency of congenital malformations was the same among those who received phen.o.barbital monotherapy and those who received monotherapy with other anticonvulsants (Bertollini et al et al., 1987). A slight, but significant, reduction in birth weight and head circ.u.mference was found among 55 newborns born to epileptic women who used phen.o.barbital during gestation, compared to newborns of women without epilepsy (Mastroiacovo et al et al., 1988). Notably, a similar effect on head circ.u.mference was observed among the newborns of women with epilepsy who received no treatment, implicating the disease. No increased frequency of congenital malformations was found among the offspring of over 1400 pregnant women who received phen.o.barbital during the first trimester (Heinonen et al et al., 1977).

Box 10.3 Commonly used sedatives, hypnotics, and tranquilizers Barbiturates Barbiturates Oxazepam (Serex) Amobarbital (Amytal) Miscellaneous sedatives and hypnotics Aprobarbital (Alurate) Chloral hydrate Butalbital Eszopiclone (Lunesta) Mephobarbital (Mebaral) Ethchlorvynol (Placidyl) Pentobarbital (Nembutal) GHB gammahydroxy butyrate Phen.o.barbital Hydroxyzine (Vistaril, Atarax) Secobarbital (Seconal) Meprobamate (Equanil, Equagesic, Miltown, Benzodiazepines Deprol) Alprazolam (Xanax) Methaqualone (Quaalude, Sapor, Parest) Chlordiazepoxide (Librium) Ramelteon (Rozerem) Clonazepam (Klonopin) Zaleplon (Sonata) Diazepam (Valium) Zolipidem (Ambien) Lorazepam (Ativan) 196.

Psychotropic use during pregnancy Characteristic dysmorphic features of the fetal hydantoin syndrome are commonly seen in newborns of women with epilepsy treated with both phenytoin and phen.o.barbital during gestation, and are discussed in Chapter 9 in the section on antiseizure medications during pregnancy. Sporadic reports of similar dysmorphic features among the infants of women with epilepsy who received phen.o.barbital monotherapy have been published (Robert et al et al., 1986; Seip, 1976).

The frequency of cleft palate, cardiovascular defects, and other congenital malformations were increased among the offspring of pregnant mice or rats given phen.o.barbital in doses greater than those used in humans (Finnell et al et al., 1987; Fritz et al et al., 1976; Nishimura et al et al., 1979; Sullivan and McElhatton, 1977; Vorhees, 1983). Malformations observed included facial anomalies similar to those observed in human newborns delivered to women with epilepsy who received anticonvulsants during gestation. A decrease in the number of specific brain cells and changes in neonatal behavior have been observed in animal studies of gestational exposure to the drug (Bergman et al et al., 1980; Takagi et al et al., 1986; Vorhees, 1983, 1985). The relevance of these observations to the clinical use of phen.o.barbital in humans is unknown.

Transient neonatal sedation or withdrawal symptoms that include hyperactivity, irritability, and tremors have been observed among newborns exposed to phen.o.barbital during pregnancy (Desmond et al et al., 1972; Koch et al et al., 1985). Hemorrhagic disease of the newborn has been a.s.sociated with phen.o.barbital use during pregnancy and typically begins within the first 24 h of life (Gimovsky and Petrie, 1986; Mountain et al et al., 1970).

The exact cause of this hemorrhagic defect is unknown, but is probably related to phen.o.barbital induction of fetal liver microsomal enzymes that deplete fetal vitamin K and suppress the synthesis of vitamin-K-dependent clotting factors II, VII, IX, and X. In contrast, maternal phen.o.barbital therapy immediately before delivery has been used to prevent intraventricular hemorrhage in premature newborns (Morales and Koerten, 1986; Shankaran et al et al., 1986).

A follow-up study of 114 adult males whose mothers used phen.o.barbital while pregnant showed lowered IQ scores by approximately 710 points, and it was concluded that phen.o.barbital exposure during early development can have long-term deleterious cognitive effects. However, detrimental environmental conditions (i.e., maternal epilepsy) may intensify such negative outcomes (Reinisch et al et al., 1995).

AMOBARBITAL.

Amobarbital, a barbiturate, is an effective sedative usually administered orally. The frequency of major and minor congenital anomalies was not increased among 298 infants born to women treated with amobarbital exposure during the first trimester (Heinonen et al et al., 1977). Amobarbital use during the first trimester was possibly a.s.sociated with cardiovascular defects (seven cases), inguinal hernia (nine cases), clubfoot (four cases), genitourinary anomalies (three cases), and polydactyly in Black infants (two cases). In a survey including over 1300 women exposed to multiple agents, of whom 175 infants were exposed to amobarbital during the first trimester, the frequency of congenital anomalies was increased (Nelson and Forfar, 1971). Authorities in the field generally believe that this drug is not likely to be a teratogen and that the significant a.s.sociations may be due to chance and conducting multiple statistical comparisons (Friedman and Polifka, 2006).

Sedatives, hypnotics, and tranquilizers 197.

APROBARBITAL.

Aprobarbital is a barbiturate used as a sedative and hypnotic agent. No information has been published regarding its safety for use during pregnancy. Furthermore, no studies in animals evaluating the teratogenic effects of aprobarbital have been published.

BUTALBITAL.

A number of a.n.a.lgesic compounds contain butalbital, a short-acting barbiturate with hypnotic and sedative properties. Among 112 infants whose mothers took butalbital during the first trimester, no increased frequency of congenital anomalies was found among the offspring (Heinonen et al et al., 1977). Transient neonatal withdrawal was reported in a.s.sociation with butalbital use late in gestation (Ostrea, 1982). No animal studies of possible teratogenic effects of butalbital have been published.

PENTOBARBITAL.

Pentobarbital is an effective, short-acting barbiturate that is used as a hypnotic and sedative agent, and is typically given orally. Among 250 infants whose mothers took pentobarbital during the first trimester, the frequency of congenital malformations was not increased (Heinonen et al et al., 1977). Similarly, among more than 50 newborns born to women exposed to pentobarbital during the first trimester of gestation, the frequency of birth defects was no greater than expected (Jick et al et al., 1981).

Skeletal and craniofacial defects, as well as fetal loss, were increased among the offspring of pregnant mice, golden hamsters, and rabbits given pentobarbital many times the doses that are used in humans (Hilbelink, 1982; Johnson, 1971; Setala and Nyyssonen, 1964). Changes in behavior and decreased brainbody weight ratios were reported among the offspring of pregnant rats administered 2040 times the human dose of pentobarbital during embryogenesis (Martin et al et al., 1985). The relevance of these findings in animals to the clinical use of this barbiturate in humans is unknown.

MEPHOBARBITAL.

Mephobarbital, which is used as an anticonvulsant and sedative, is metabolized by the liver to phen.o.barbital and thus has similar properties. Results in a j.a.panese multi-inst.i.tutional study that included the frequency of congenital anomalies in a cohort of 111 infants born to pregnant epileptics who used mephobarbital during the first trimester, were similar to those for the infants of pregnant epileptics treated with other medications (Nakane et al et al., 1980). In a small case series, the frequency of congenital malformations was no greater among the newborns of 17 epileptic mothers exposed to mephobarbital during the first trimester of pregnancy than among the newborns of epileptic mothers who received no treatment (Annegers et al et al., 1974). No animal teratology studies of mephobarbital have been published.

SECOBARBITAL.

Secobarbital is as effective as pentobarbital and is generally administered orally. Among 378 infants born to women who took secobarbital during the first trimester, the frequency of congenital anomalies was not increased (Heinonen et al et al., 1977). One report of an infant with neonatal withdrawal symptoms of hyperirritability and seizures a.s.sociated 198 198 Psychotropic use during pregnancy with maternal use of large doses of secobarbital throughout gestation has been published (Bleyer and Marshall, 1972).

Benzodiazepines Benzodiazepines are minor tranquilizers with mild anticonvulsant and sedation properties (Box 10.3). These agents differ in potency and duration of effect, and indications for their use are based upon these features. They are commonly used anxiolytic agents. Of these, diazepam is probably the most frequently used drug.

DIAZEPAM.

Diazepam is a benzodiazepine used as a tranquilizer, skeletal muscle relaxant, and preanesthetic medication. Diazepam is also used to treat alcohol withdrawal and as an adjunct to anticonvulsants in the treatment of seizure disorders. It is the most extensively used drug in the benzodiazepine family.

Inconsistencies in the currently available epidemiological data on the risk of congenital anomalies among newborns of women who were exposed to diazepam during gestation, confound the issue. Diazepam use during the first trimester was not a.s.sociated with an increased frequency of malformations among the newborns of more than 150 women in two cohorts, or among 60 newborns of women who used the drug in the first trimester (Aselton et al et al., 1985; Crombie et al et al., 1975; Jick et al et al., 1981).

In contrast, first-trimester diazepam use was increased almost threefold among 1427 infants with congenital malformations compared to controls in one study (Bracken and Holford, 1981), but not in another casecontrol study that included 417 newborns with multiple congenital malformations (Czeizel, 1988). A hypothesized 'benzodiazepine embryofetopathy' (typical facial features, neurological dysfunction, and other anomalies) (Laegreid et al et al., 1987, 1989) is not widely accepted in the human teratology community.

Some early evidence suggested that maternal use of diazepam or other benzodiazepines during the first trimester of gestation was a.s.sociated with facial clefts in the infants (Aarskog, 1975; Safra and Oakley, 1975; Saxen, 1975; Saxen and Saxen, 1975); more extensive studies have not confirmed this a.s.sociation. Among the infants of women who had first-trimester exposure to antineurotics (mainly diazepam) the frequency of congenital anomalies was not increased (Crombie et al et al., 1975; Rosenberg et et al al., 1983; Shiono and Mills, 1984). On balance, the possible risk of cleft lip or palate in the infant of a women exposed to diazepam during the first trimester, if increased at all, is less than 1 percent. Notably, family history of congenital anomalies is a confounder in at least two of these studies.

Maternal use of diazepam or related compounds during the first trimester of pregnancy and an increased risk for cardiovascular anomalies was observed in two casecontrol studies involving 773 infants (Bracken and Holford, 1981; Rothman et al et al., 1979). However, in a rea.n.a.lysis, no significant a.s.sociation was found (Bracken, 1986).

In a follow-up study of 298 infants with congenital heart defects, no a.s.sociation with first-trimester diazepam was found (Zierler and Rothman, 1985). The risk for congenital heart disease among the infants of women who have first-trimester exposure to diazepam, is probably not increased, but if it is increased the magnitude is small (< 12="">

Sedatives, hypnotics, and tranquilizers 199.

Diazepam is readily transferred across the placenta to the human fetus and becomes concentrated in the fetal compartment with a 2:1 ratio (Erkkola et al et al., 1974). Apnea, hypotonia, and hypothermia were observed in newborns of women who took diazepam during the third trimester of pregnancy or peripartum (Cree et al et al., 1973; Gillberg, 1977; Owen et al et al., 1972; Speight, 1977). Tremors, irritability, and hypertonia similar to neonatal narcotic withdrawal was observed in some infants chronically exposed in utero in utero during the third trimester to diazepam (Rementeria and Bhatt, 1977). Loss of beat-to-beat fetal heart rate variability was a.s.sociated with diazepam exposure during late pregnancy (Scher during the third trimester to diazepam (Rementeria and Bhatt, 1977). Loss of beat-to-beat fetal heart rate variability was a.s.sociated with diazepam exposure during late pregnancy (Scher et al et al., 1972) and decrease in fetal movement (Birger et al et al., 1980). The effect of prenatal exposure to this drug and any untoward central nervous system function in later life is unknown. A few reports of normal infants born to women who took toxic doses of diazepam during gestation, with the majority of cases occurring after the first trimester have been published (Cerqueira et al et al., 1988; Czeizel, 1988).

Animal studies indicate that diazepam is a teratogen in mice and hamsters, but only when doses hundreds of times greater than those used in humans are administered (Kellogg, 1988; Weber, 1985).

CHLORDIAZEPOXIDE.

Chlordiazepoxide is a benzodiazepine tranquilizer that has less potency than diazepam on a milligram basis. It also has less anticonvulsant, muscle relaxant and sedative properties, but is effective in treating alcohol withdrawal. In several cohort studies, the frequency of congenital defects was not increased among more than 480 newborns whose mothers had first-trimester exposure to chlordiazepoxide (Crombie et al et al., 1975; Hartz et al et al., 1975; Heinonen et al et al., 1977; Kullander and Kallen, 1976a, 1976b). Two casecontrol studies reported no a.s.sociation between maternal use of this benzodiazepine in the first trimester and congenital defects (Bracken and Holford, 1981; Rothman et al et al., 1979). In contrast, in a small cohort study an a.s.sociation was reported between congenital malformations in 35 infants and maternal use of chlordiazepoxide in the first 42 days of gestation (Milkovich and van den Berg, 1974). However, there was no pattern to the anomalies observed.

Maternal chlordiazepoxide was a.s.sociated with neonatal withdrawal beginning on day 26 of life (Athinarayanan et al et al., 1976), which is an unusually long lag time for such an effect. Withdrawal symptoms included extreme tremulousness and irritability.

Chlordiazepoxide given to pregnant hamsters in doses greater than those used in humans during embryogenesis resulted in an increased frequency (dose-dependent) of central nervous system anomalies and maternal toxicity (Guram et al et al., 1982). The frequency of congenital anomalies was not increased among the offspring of pregnant rats given benzodiazepine, but fetal loss, growth r.e.t.a.r.dation, and skeletal variants were increased in frequency with higher doses (b.u.t.tar, 1980; Saito et al et al., 1984).

LORAZEPAM.

Lorazepam is a benzodiazepine cla.s.s minor tranquilizer. The drug is used as an antianxiolytic and hypnotic drug. It is also used as a preanesthetic medication because of its amnesic action. In a casecontrol study, a significant a.s.sociation between first-trimester exposure to lorazepam and a.n.a.l atresia was published (Bonnot et al et al., 1999), but the meaning of the a.s.sociation is unclear. Placental transfer of lorazepam was reported by several investigators (de Groot et al et al., 1975; Kanto et al et al., 1980; McBride et al et al., 1979).

200.

Psychotropic use during pregnancy Transient neonatal hypotonia was observed in newborns of women who took lorazepam late in gestation, either chronically or intrapartum (McAuley et al et al., 1982; Whitelaw et al et al., 1981). Intravenously administered maternal lorazepam in hypertensive gravidas was a.s.sociated with low Apgar scores, hypothermia, poor feeding, and a requirement for a.s.sisted ventilation in the infant. No congenital anomalies were reported among the offspring of pregnant rats and mice given up to 4 mg/kg.day lorazepam during organogenesis (Esaki et al et al., 1975).

ALPRAZOLAM.

Alprazolam is a benzodiazepine tranquilizer. The rate of malformations was approximately 5 percent in over 400 births reported to the manufacturer (St Clair and Schirmer, 1992). In a casecontrol study from Hungary, the a.s.sociation between alprazolam exposure and congenital anomalies was slightly elevated, but not significantly so (Eros et al et al., 2002). Several series containing more than 300 pregnancies followed through teratogen information services found no apparent increase in congenital anomalies (Friedman and Polifka, 2006) This agent, as other benzodiazepines, may cause hypotonia and hypothermia in the newborn (Yonkers and Cunningham, 1993).

OTHER BENZODIAZEPINES.

Oxazepam and clonazepam are benzodiazepine tranquilizers. Among 89 infants born to women who used oxazepam during the first trimester, there were no congenital anomalies (Ornoy et al et al., 1998). Small numbers of first-trimester exposure to clonazepam are published in clinical case series, but they are confounded by concomitant use of other known teratogens (anticonvulsants), as well as small sample sizes and sample selection bias (Friedman and Polifka, 2006).

Congenital anomalies were not increased in frequency among the offspring of pregnant rabbits or rats administered oxazepam in doses greater than those used in humans (Owen et al et al., 1970; Saito et al et al., 1984). Changes in behavior were observed among the offspring of pregnant mice given oxazepam in doses four to 42 times those used clinically (Alleva and Bignami, 1986).

MISCELLANEOUS.

Hydroxyzine Hydroxyzine is a piperazine antihistaminic compound that is used to treat anxiety, pruritus, nausea, and vomiting. The frequency of congenital anomalies in a double-blind controlled study was not increased among 74 newborns exposed in utero in utero to hydroxyzine (50 mg/day) during the first trimester (Erez to hydroxyzine (50 mg/day) during the first trimester (Erez et al et al., 1971). Birth defects were not increased in frequency among 50 infants born to women who used hydroxyzine during the first trimester (Heinonen et al et al., 1977). Hydroxyzine has been shown to be a teratogen in rats (Giurgea and Puigdevall, 1968; King and Howell, 1966).

Chloral hydrate Chloral hydrate is an effective hypnotic and sedative agent. There is a paucity of information regarding the safety of chloral hydrate use during pregnancy. However, among Miscellaneous Miscellaneous 201.

71 infants born to women who used chloral hydrate during the first trimester, the frequency of congenital anomalies was not increased (Heinonen et al et al., 1977). No gross external defects were observed in pregnant mice with chloral hydrate in doses less than one to five times the human dose (Kallman et al et al., 1984).

Ethchlorvynol Ethchlorvynol is a tertiary acetylenic alcohol and is used as an oral hypnotic and sedative agent. No studies have been published regarding the frequency of congenital malformations among newborns of women exposed to ethchlorvynol during gestation.

Symptoms of neonatal withdrawal were observed in the newborn of a woman who was treated with ethchlorvynol as a hypnotic during the last 3 months of gestation. Neonatal withdrawal symptoms observed were jitteriness, irritability, and hypotonia (Rumak and Walravens, 1973). No animal studies evaluating the teratogenic effects of ethchlorvynol are published, but behavioral changes were observed among the offspring of pregnant rats treated with ethchlorvynol in doses greater than those used in humans (Peters and Hudson, 1981).

Meprobamate Meprobamate is a carbamate tranquilizer that is useful in the treatment of anxiety but seems to be less effective than the benzodiazepines. The most common side effect is drowsiness. Inconsistencies in studies of the possible teratogenic effects of meprobamate in humans make it difficult to a.s.sess the risk of congenital anomalies with exposure to the drug in therapeutic doses during embryogenesis. Reports of an a.s.sociation between maternal use of this drug during the first trimester of pregnancy and a variety of congenital defects in newborns have been published, but the a.s.sociation is weak, and in no two studies was the same defect present. Among 66 infants born to women exposed to meprobamate in the first 42 days after their last menstrual period, congenital anomalies were increased fourfold (Milkovich and van den Berg, 1974). No apparent pattern of congenital anomalies was identified, but there were five infants with congenital heart disease. The frequency of hypospadias was increased among the 186 male infants born to women treated with meprobamate during the first trimester of pregnancy (Heinonen et et al al., 1977), but the finding was disregarded because of the small sample size. Accordingly, the relationship is probably a random finding, not representing a causal link. A third study had an increased frequency of major congenital anomalies among the newborns of more than 50 pregnant women given meprobamate during the first trimester (Jick et al et al., 1981), but no other details are available. Other studies have failed to find an a.s.sociation between the first-trimester use of meprobamate and congenital malformations. Among 356 pregnant women given meprobamate during the first trimester, the frequency of congenital anomalies was not increased (Heinonen et al et al., 1977). Another cohort study of congenital anomalies among 207 infants whose mothers used meprobamate during the first trimester failed to find an a.s.sociation (Belafsky et al et al., 1969). However, it should be noted that these studies a.n.a.lyzed only therapeutic dose exposures for infants examined for birth defects.

202.

Psychotropic use during pregnancy Table 10.4 Teratogen Information System (TERIS) and Food and Drug Administration (FDA) pregnancy risk ratings Teratogen Information System (TERIS) and Food and Drug Administration (FDA) pregnancy risk ratings Drugs TERIS risk FDA Pregnancy risk rating Alprazolam Unlikely Dm Amitriptyline Unlikely Cm Amobarbital None to minimal D*

Amoxapine Undetermined Cm Aprobarbital Undetermined C.

Bupropion Undetermined Bm Butalbital Unlikely C*

Chloral hydrate Unlikely Cm Chlordiazepoxide Unlikely D.

Chlorpromazine Unlikely C.

Clomipramine Unlikely Cm Clonazepam Undetermined Dm Clozapine Undetermined Bm Desipramine Undetermined C.

Diazepam Minimal D.

Doxepin Undetermined C.

Ethchlorvynol Undetermined Cm Fluoxetine Unlikely Cm Fluphen.a.z.ine Unlikely C.

Haloperidol Unlikely Cm Hydroxyzine Unlikely C.

Imipramine Unlikely D.

Isocarboxazid Undetermined C.

Lithium Small D.

Lorazepam Undetermined Dm Loxapine Unlikely C.

Maprotiline Undetermined Bm Mephobarbital Unlikely Dm Meprobamate Minimal D.

Mesoridazine Undetermined C.

Molindone Undetermined C.

Nortriptyline Undetermined D.

Oxazepam Unlikely D.

Pentobarbital Unlikely Dm Perphen.a.z.ine Unlikely C.

Phenelzine Undetermined C.

Phen.o.barbital Chronic anticonvulsive D.

Phenytoin/fosphenytoin Small to moderate D.

Protriptyline Undetermined C.

Secobarbital None Dm Sertraline Unlikely Bm Thioridazine Unlikely C.

Thiothixene Undetermined C.

Tranylcypromine Undetermined C.

Trazodone Unlikely Cm Trifluoperazine Unlikely C.

Compiled from: Friedman et al., Obstet Gynecol 1990; 75 75: 594; Briggs et al., 2005; Friedman and Polifka, 2006.

Miscellaneous 203.

The frequency of birth defects was not increased among the offspring of pregnant mice, rats, or rabbits given meprobamate in doses greater than those used in humans (range 2.516 times) (Brar, 1969; Clavert, 1963; Werboff and Dembicki, 1962). In other studies when doses 16 times greater than those used in humans were administered to rabbits, and doses 2.5 or 20 times the typical dose in humans were given to pregnant rats, fetal and neonatal loss was increased (Bertrand, 1960; Clavert, 1963; Werboff and Kesner, 1963).

Methaqualone Methaqualone (Quaalude, Sopor, Parest) is an effective hypnotic and sedative agent and is not presently commercially available. No clear medicinal advantage of methaqualone over the other available hypnotics can be shown and the drug is commonly abused by drug-dependent people. Tolerance to the drug develops in abusers. No published reports are available that a.n.a.lyze the possible a.s.sociation of the use of methaqualone during pregnancy with congenital malformations. However, its use during gestation is not recommended because of its abuse potential. The frequency of congenital anomalies was not increased among rats or rabbits whose mothers were administered 200 mg/kg methaqualone orally (rabbits) from days 1 to 29 or 100 mg/kg (rats) from days 1 to 20 (Bough et al et al., 1963) (Table 10.4).

Electroconvulsive therapy High-voltage electrical shock is used to treat some psychiatric disorders, although it may also occur in accidental electrocution. The mechanism of action of electroconvulsive therapy is unknown. However, it is clearly understood that the seizure produced by electroconvulsive therapy is necessary for therapeutic efficacy (Ottosson, 1962a, 1962b). Electroconvulsive therapy was used safely in the treatment of depression in a pregnant woman following expanded clinical guidelines that included the presence of an obstetrician during treatment, endotracheal intubation, low-voltage, nondominant therapy with electrocardiographic and electroencephalogram monitoring, Doppler ultrasonography of fetal heart rate, tocodynamometer recording of uterine tone, arterial blood gases during and after treatment, glycopyrrolate (anticholinergic of choice) use during anesthesia, and weekly nonstress tests (Wise et al et al., 1984). The frequency of birth defects among the newborns of 318 women who received electroconvulsive therapy during gestation has not increased (Impastato et al et al., 1964).

Reports of uterine contractions, v.a.g.i.n.al bleeding, and transient benign fetal cardiac arrhythmias have been published (Miller, 1994b; Rabheru, 2001). Miscarriage was reported following a third electroconvulsive therapy session in the first trimester of pregnancy (Moreno et al et al., 1998). One infant was described with hydrops fetalis and meconium peritonitis after the mother received electroconvulsive therapy during the third trimester of pregnancy (Gilot et al et al., 1999). As is usually the case with isolated reports, it is not possible to evaluate any causal links with anecdotal data.

No animal studies evaluating the teratogenic effects of high-voltage electrical shock have been published.

204.

Psychotropic use during pregnancy SPECIFIC CONDITIONS.

Depression Management of depression during pregnancy should be undertaken in consultation with a psychiatrist and/or psychologist (Yonkers, 2003). Although psychotherapy or hospitalization in a supportive environment is the first consideration in treatment (Spinelli, 2001; Yolles, 2001), antidepressant therapy may be necessary if these regimens are unsuccessful (Yolles, 2001; Robinson et al et al., 1986). Indeed, antidepressant medications are indicated in the pregnant woman whose depression is so severe that it threatens her life and the life of her unborn child (Yolles, 2001; Yonkers 2003). Since the medications used in the treatment of depression have potential fetal risks and may result in obstetric complications and long-lasting sequelae, the minimal effective antidepressant dose should be initiated and maintained. Most antidepressants have established therapeutic serum levels that can be monitored.

No antidepressant has proven safety for use during gestation, although some are better studied than others. Thus the selection of an antidepressant is dependent upon a patient's past response, side effects, and potential teratogenic effects of the particular agent. However, it is recommended that one uses an agent that has been relatively well studied during pregnancy and that has relatively few side effects (Miller, 1994a, 1996).