Illinois Teratogen Information Service

1-800-252-4847

Gastroesophogeal Reflux (GERD) Medications in Pregnancy

Gastroesophogeal Reflux (GERD) Medications in Pregnancy

Posted by admin on June 2nd, 1999 — in newsletter

PDF Version 

Gastroesophogeal Reflux (GERD) Medications in Pregnancy

Deborah Lilienthal, BS; Kelly Ormond, MS, CGC, Eugene Pergament, MD, PhD

Volume 7(5), June 1999

Gastroesophageal reflux disease (GERD) is the movement of acidic gastric contents into the lower esophagus. Often described as heartburn or “acid indigestion,” reflux occurs in approximately two thirds of all pregnancies. Almost everyone experiences gastroesophogeal reflux at some time, although in some individuals reflux is frequent or severe enough to require daily medication. There is also a growing percentage of women with GERD prior to pregnancy who will require maintenance medications throughout gestation. The origin of GERD is multifactorial, but the predominant factor is a decrease in lower esophageal sphincter pressure. Since mechanical factors and female sex hormones, especially progesterone, play a role in GERD (Baron and Richter, 1992), heartburn is more common during the last months of pregnancy, when the growing fetus presses against the stomach and hormones are at high levels. This RISK//NEWSLETTER will present information on the reproductive risks of GERD medications.

Mild heartburn may be treated with lifestyle and dietary modifications. Good posture while eating and avoiding lying down after meals may reduce the occurrence of heartburn. Smoking should also be avoided because it increases stomach acidity.

Antacids

Antacids are commonly used by many women during pregnancy. Available without a prescription, they neutralize stomach acid to provide temporary relief. A variety of antacids are available. The most common antacids on the market are: Mylanta (calcium carbonate and magnesium hydroxide); Maalox (magnesium hydroxide and aluminum hydroxide); Tums (calcium carbonate); and Rolaids (calcium carbonate and magnesium hydroxide). Although there is little empiric data on human exposure to antacids, they are a common exposure in pregnancy, and as such are unlikely to significantly increase the risk for birth defects at therapeutic doses.

Aluminum Hydroxide

Animal studies of aluminum hydroxide at oral doses up to 64 times the human dose did not show significant maternal or developmental toxicity (Domingo et al., 1989; Gomez et al., 1990; Colomina et al., 1994). Therefore, at usual human doses, aluminum hydroxide is not likely to increase the risk for birth defects.

Calcium Carbonate

In an animal study done by Shackelford et al. (1993), rats were given up to 1.25% dietary calcium carbonate before mating and during organogenesis. Dietary calcium was not found to have adverse effects of the offspring. Severe hypercalcemia, a potentially life-threatening condition, was reported in a pregnant woman with excessive ingestion of absorbable calcium antacid. She was treated and the child was delivered a month later with an uncomplicated neonatal course (Kleinman et al., 1991). Even with toxic levels of calcium carbonate in the mother, teratogenic effects were not seen in the fetus. Therefore, calcium carbonate is unlikely to significantly increase the risk for birth defects.

Magnesium Hydroxide

In one study, magnesium hydroxide was administered to 27 pregnant women in the third trimester without adverse effects (Rudnicki et al., 1991). Although limited information is available, this data does not support an association between magnesium hydroxide and birth defects.

Histamine-2 Receptor Blockers (H2RA)

Histamine-2 Receptor Blockers (H2RA) treat the discomfort of heartburn and acid indigestion by blocking histamine, which decreases acid secretion. They are also commonly used during pregnancy, and they are available both over the counter and by prescription at higher doses. The majority of reproductive information for H2RAs regards cimetidine and ranitidine.

Cimetidine (Tagament)

Cimetidine has weak antiandrogenic effects in animals and antiandrogenic effects in humans, causing impotence and oligospermia (Sawyer et al., 1981). While no studies have examined these possible antiandrogenic effects in human pregnancy, two studies found that cimetidine adversely affects male androgenization and neuroendrocrine programming in rats (Anand and Van Theil, 1982; Parker et al., 1984). It is theoretically possible that use in pregnant women may adversely affect adult sexual behavior and development of male progeny. Other animal studies showed no difference in masculinity between those animals taking cimetidine from their controls (Hoie et al., 1994; Shapiro and Bitar, 1991; Shapiro et al., 1988; Walker et al., 1987). It remains unclear what significance, if any, these findings may have on human exposures.

A prospective study of 10 women exposed to cimetidine in the first trimester reported 2 therapeutic abortions and 8 normal births (Koren and Zemlickis, 1991). A retrospective study of 460 newborns exposed to cimetidine in the first trimester found no increase in major birth defects, but slightly more heart defects were observed than expected (Briggs et al., 1998). Other studies have not shown an increase in heart defects. The manufacturer reported three newborns with congenital birth defects (congenital heart defect, clubfoot, and mental retardation, respectively) after in utero exposure to cimetidine. No pattern was seen in these infants and the defects were not attributed to the use of cimetidine (Briggs et al., 1998). While limited, this data does not suggest an association between cimetidine and birth defects.

Cimetidine is also used at term to prevent maternal Mendelson’s syndrome. There are several reports of women exposed late in pregnancy, without reported teratogenic effects in the newborns (e.g., Carazza et al., 1982). Glade et al. (1980) reports transient neonatal liver toxicity in a newborn exposed to cimetidine at term. This has not been reported in any subsequent studies.

Famotidine (Pepcid)

Animal studies have shown no adverse effects when given famotidine at doses as high as 2,000 mg/kg/day, much higher than the recommended human dose (Shibata et al., 1983; Burek et al., 1985). There is little information in the medical literature on the effects of famotidine in human pregnancy. In a retrospective study of 33 newborns exposed to famotidine in the first trimester, 2 major birth defects were seen; there was no pattern to these defects(Briggs et al., 1998). The number of exposures is too small to make a risk assessment. Famotidine does not have antiandrogenic effects.

Nizatidine (Axid)

Nizatidine was given orally to rats and rabbits at doses as high as 1500mg/kg/day and no teratogenic effects were seen (Morton , 1987). In contrast, the manufacturer reports animal studies with congenital malformations at doses of 20mg/kg and 50mg/kg. These malformations included cardiac defects, neural tube defects and cutaneous edema (Briggs et al., 1998). There is no human epidemiologic data on the effects of nizatidine during pregnancy, and its risk is therefore undetermined. One case report of a woman exposed to nizatidine in the second trimester shows that she delivered a healthy infant (Briggs et al., 1998). Unlike cimetidine, nizatidine is not an androgen antagonist (Neubauer et al., 1990).

Ranitidine (Zantac)

Animal studies on ranitidine have not shown an increased risk for malformations. A prospective study of 13 women exposed to ranitidine during the first trimester reported 10 normal births, 2 spontaneous abortions and one infant born with a hemangioma on the right upper eyelid (Koren and Zemlickis, 1991). A retrospective study of 516 newborns exposed to ranitidine in the first trimester reported no increase in malformations nor pattern to those noted (Briggs et al., 1998).

Most data on ranitidine is regarding use near delivery to prevent Mendelson’s syndrome. There have not been reports of adverse effects in newborns exposed to ranitidine near delivery (summarized in Briggs et al., 1998). While limited, the data does not support an association between the drug and congenital defects. Antiandrogenic effects have not been seen with the use of ranitidine (Parker et al.,1984).

Proton Pump Inhibitors (PPIs)

Proton pump inhibitors decrease the stomach’s production of acid more completely than the H2RA’s by stopping the stomach’s acid pump, which is the final step of acid secretion. PPIs are a relatively new class of GERD treatment, and as such, less information is available.

Lansoprazole (Prevacid)

Animal studies on lansoprazole in rabbits and rats did not find evidence that lansoprazole impairs fertility or teratogenicity at 16 to 80 times the human doses, respectively (Schardein et al., 1990; Briggs et al., 1998). There have been no reports on the effects of lansoprazole use during human pregnancy, and as such its risk is undetermined.

Omeprazole (Prilosec)

Animals given up to 345 times the recommended human dose of omeprazole did not show teratogenic effects, although there was a slight increase in miscarriage and fetal mortality (Briggs et al., 1990). A case report of a woman who used omeprazole in three pregnancies, including one in the first trimester, showed that she delivered three healthy infants (Harper et al., 1995). Several case reports exist of adverse outcomes after omeprazole use. The FDA has received 11 voluntary reports of birth defects following pregnancy exposure to omeprazole use, including four cases of anencephaly and one case of hydranencephaly after use in the second trimester (Briggs et al., 1998). Tsirigotis et al. (1995) describes a woman who ingested 20mg of omeprazole daily during two consecutive pregnancies and subsequently terminated them because of anencephaly and clubfoot, respectively. While these case reports of anencephaly suggest a pattern of defects, without background information on these pregnancies, the potential confounding factors inherent in case reports make it difficult to attribute the cause to omeprazole.

In a recent prospective cohort study of 113 women exposed to omeprazole, 101 throughout organogenesis (89%) and 15% throughout pregnancy, no association was found between in utero exposure and malformations, birth weight, gestational age at delivery, preterm deliveries, or neonatal complications (Lalkin et al., 1998). Although the case reports may be concerning, the lack of teratogenicity in animals and the recent prospective human studies show that omeprazole is unlikely to significantly increase the risk for birth defects.

Prokinetic Agents

Prokinetic agents hasten emptying of the stomach contents, resulting in less acid secretion available for reflux. Some agents also increase the “tone” of the lower esophageal sphincter, making it more difficult to open.

Cisapride (Propulsid)

Animal studies in rats show impaired fertility at 25 times the human dose. At 12 to 100 times the human dose in rats and rabbits, respectively, an increase in IUGR and neonatal death was noted (Briggs et al., 1998). In a prospective study, 129 pregnant women were exposed to cisapride, including 88 during organogenesis. There were no differences in birthweight, gestational age at delivery, and rates of livebirths, spontaneous abortions, fetal distress, and major or minor malformations among those exposed to the drug and those used in the control group. This suggests that cisapride is not likely to pose a significant teratogenic risk (Bailey et al., 1997).

Metoclopramide (Reglan)

Manufacturer’s information on mice, rats, and rabbits given doses up to 250 times the human dose, showed no evidence of fetal harm (Briggs et al., 1998). In a retrospective study of 192 newborns exposed to metoclopramide in the first trimester, 10 (5.2%) major birth defects were seen (Briggs et al., 1998). In a study by Nageotte et al. (1996) 80 women with hyperemesis used metoclopramide during pregnancy. Three women who used metoclopromide in the second trimester delivered infants with birth defects; there was no pattern to the defects, making it even less likely that metoclopramide was a causal factor. Five case reports of women exposed to metoclopramide in early pregnancy did not show teratogenic effects (Briggs et al., 1998).

Summary

Little empiric data is available on most GERD treatments in pregnancy, despite their frequent exposure. The lack of human studies on GERD medications in pregnancy makes it difficult to provide an accurate teratogenic risk assessment. As with any medication, the lowest possible dose should be taken to relieve discomfort. Pregnant women with GERD should speak to their doctors about the best treatment for their GERD, and consider weighing the need for treatment with the options and information available on use during pregnancy.

References

Anand S, VanThiel D (1982) Science 18(4571):493-4

Bailey B et al. (1997) Dig Dis Sci 42(9):1848-52.

Baron TH, Richter JE (1992) GastroenterologyClin NA 21(4):777-91.

Briggs GG et al. (1998) Drugs in Pregnancy and Lactation, 5th ed. Baltimore:Williams & Wilkins.

Burek JD et al. (1985) Digestion 32(Sup1): 7-14.

Colomina MT et al. (1994) Pharmacology & Toxicology 74(4-5):236-9.

Corazza GR et al. (1982) Clin Trials J 19:91-3.

Domingo JL et al. (1989) Life Sciences 45(3):243-7.

Koren G, Zemlickis D (1991) Am J Perinat 8(1):37-8.

Glade G et al. (1980) Am J of Dis Child 34(1):87-8.

Gomez M et al. (1990) Vet Hum Toxicol 32(6):545-8.

Harper MA et al. (1995) Am J Obstet Gynecol

173(3 Pt 1):863-4.

Health Informatics International, Inc

www.healthinformatics.com/docs/english/index.asp

Hoie EB et al. (1994) J Pharm Sciences 83(1):107-9.

Kleinman GE et al. (1991) Obstet Gyn 78(3Pt2):496-9.

Lalkin A et al. (1998) Am J Obstet Gynecol 179(3 Pt 1):727-30.

Morton DM (1987) Scand J Gasteroent 22(S 136):1-8.

Neubauer BL et al. (1990) Tox AppPharm 102:219-32.

Parker S et al. (1984) Neuro Tox & Terat. 6(4):313-8.

Parker S et al (1994) Gastroenterology 86(4):675-80.

Rudnicki M et al (1991) Acta Obstet Gyn Scand 70(6):445-50.

Sawyer D et al. (1981) Am JHospPharm 38(2):188-97.

Shackelford ME et al. (1993) Food & Chemical Toxicology 31(12):953-61.

Schardein JL et al. (1990) Yakuri To Chiryo 18(S10):119-29.

Shapiro BH, Bitar MS (1991) Tox Letters. 55(1):85-98.

Shapiro BH et al. (1988) Tox Letters 44(3):315-29.

Shibata M et al. (1983) Oyo Yakuri 26: 489-497, 543-578, 831-840.

Tsirigotis M et al. (1995) Human Repro 10(8):2177-8.

Walker TF et al. (1987) Fund Appl Tox. 8(2):188-97.

Update: Benzodiazepines in Pregnancy

Posted by admin on June 1st, 1999 — in newsletter

PDF Version 

Update: Benzodiazepines in Pregnancy

Vol. 7., No. 4 June 1999

Kelly Ormond, MS, CGC and Eugene Pergament, MD, PhD

Benzodiazepines (BZDs) are frequently prescribed during pregnancy to treat anxiety or panic disorder. As a class, the BZDs are central nervous system depressants that have anxiolytic, sedative, hypnotic, muscle relaxant and occasionally anti-epileptic properties. They are well absorbed in the body and cross the placenta easily (Kanto, 1982); elimination half-lives for the BZDs range significantly (8-48 hours), with active metabolites often remaining in the body for longer periods of time (Gilman, 1990).

Initial concern regarding BZD exposure in pregnancy arose because they act upon GABA receptors; GABA is an amino acid neurotransmitter that may be related to palatal development (Kellogg, 1988). Early studies on Valium (diazepam), a commonly prescribed BZD, showed an increased risk for oral clefting in both animals (Zimmerman, 1984) and in retrospective and case-control studies in humans (Saxon and Saxen, 1975; Safra and Oakley, 1975). This has, however, been contradicted by several recent prospective and case-controlled studies and a meta-analysis that all uniformly found no association between diazepam use and clefting (Altshuler et al., 1996; Bracken, 1986; Czeizel, 1988; Ornoy et al., 1998; Pastuszak et al., 1994; Rosenberg et al., 1983; Shiono and Mills, 1984). In recent years, several prospective studies have addressed the potential teratogenicity of multiple BZDs. The association between BZDs and clefting, and birth defects in general, remains unclear, and it will be reviewed in more detail in this RISK//NEWSLETTER. BZDs were reviewed in the September, 1995 (RISK//NEWSLETTER 4(2)); this newsletter serves as an adjunct to that issue.

General studies on BZDs

Many of the studies on BZD exposure in pregnancy have lumped various drugs into a single analysis, making it difficult to determine if specific medications pose teratogenic risk in pregnancy. McElhatton (1994) provides an excellent review of many of the studies on BZDs. While there have been mixed findings, these studies do not suggest an overall increase in malformations after in utero exposure to BZDs. Pastuszak et al. (1994) prospectively ascertained 137 women exposed to BZDs, primarily diazepam (N=43) and lorazepam (N=33) and found no differences from control groups in frequency of malformations or miscarriage, birth weight, gestational age or measures of the Denver developmental scale at various ages. Johnson et al. (1995) presented an abstract reporting 272 women exposed to alprazolam, lorazepam and clonazepam; of the 186 liveborns, 15 had malformations, including four cardiac defects and six inguinal hernias, which the authors speculated may be secondary to the muscle relaxant properties of BZDs. In a separate abstract, Godet et al. (1995) reported on 187 malformed infants exposed to BZDs; while no anomaly was more frequent in this group as compared to a control group of over 10,000 malformed infants, associations between lorazepam and anal atresia and between bromazepam and urinary anomalies were noted, but there was no association between clefting and any BZD. Most recently, Ornoy et al. (1998) prospectively reported 460 pregnancies exposed to BZDs in the first trimester and found no increase in birth weight, gestational age or malformations, but more cardiac defects were present in the exposed versus control group; a slight increase in miscarriage and induced abortions was also observed.

Aside from the concerns regarding malformations after in utero exposure to BZDs, there are reports of transient neonatal withdrawal symptoms and even of a possible syndrome of BZD exposure after significant maternal exposure (Laegrid et al., 1989; Bergman et al., 1992). Laegrid et al. (1989) described facial dysmorphology similar to fetal alcohol syndrome, involving impaired growth, hypotonia, developmental and motor delays and transient neonatal withdrawal after significant maternal BZD exposures. Other studies have failed to support this association, suggesting that perhaps chronic use at high dosages is required to produce this syndrome. There has also been controversy about whether an autosomal recessive condition (Zellweger syndrome) explains some of the physical and developmental features noted. The confounding effect of other maternal drug use in Laegrid’s studies has also been raised. Most recently, the Michigan Medicaid study (Bergman et al., 1992) looked at 80 women who filled over 10 benzodiazepine prescriptions during a pregnancy and saw no increase in malformations among those exposed in utero. Concurrent maternal alcohol and substance exposure in pregnancy significantly biases all of the above studies. While no long term studies have been performed to assess neurodevelopment in exposed children, the potential exists for neurobehavioral teratogenicity after exposure to BZDs, and this has been noted in animal studies (Schardein, 1993).

Reproductive Data on Specific BZDs

Alprazolam (Xanax)

Several human studies exist on alprazolam exposure in pregnancy. Postmarketing research of 411 women with first trimester exposure to alprazolam did not suggest an increased frequency of malformations (St. Clair et al., 1992). Separate prospective studies of 133 and 149 women, respectively, found no increased risk of malformations nor any pattern to the malformations described (Johnson et al., 1995; Ornoy, 1998). Neonatal withdrawal symptoms have been noted after exposure to alprazolam in late pregnancy (Barry and St.Clair, 1987) and breast-feeding (Anderson and McGuire, 1989). Alprazolam has a relatively short half-life (<12 hours) compared to other BZDs.

Chlordiazepoxide (Librium)

Data on chlordiazepoxide exposure in pregnancy has been contradictory. 175 pregnancies exposed to chlordiazepoxide showed an increased frequency of malformations (Milkovich, 1974). In contrast, two studies of 257 and 136 women using Librium in the first trimester found no increase in malformations (Hartz, 1975; Crombie, 1975). This was supported by a large retrospective study of 788 women (Rosa, cited in Briggs, 1998) that also showed no increase in malformations, but there was a slight increase in cardiac anomalies (10 vs. 7 expected). A case-control study of infants with cardiac defects also showed a slight association with chlordiazepoxide exposure (Rothman et al., 1979), but it is unclear what to make of this association. Animal studies involving rats and mice show no increase in malformations at doses lower than the maternal toxicity levels; however, there is some evidence of low birth weight and behavioral changes at doses 2-7X the human dose. Withdrawal symptoms have been noted after exposure to chlordiazepoxide near term (Briggs, 1998).

Clonazepam (Klonopin)

Clonazepam has not been shown to increase malformations in rats or rabbits. A single retrospective study of 19 women exposed in the first trimester showed 3 malformations (including 2 heart defects); because of the small study size, the implications of this finding are unclear (Briggs, 1998). Prospective studies of 60 and 69 women each found a slight increase in malformations, but given the small number of women in each study, it remains difficult to determine causality from these studies (Johnson et al., 1995; Ornoy, 1998). One complicating factor is that clonazepam is also used to treat seizure disorder and therefore, an increase in malformations may be due to epilepsy rather than medication use. A case control study of anti-epileptic medications, including clonazepam, did not show an association with malformed infants and clonazepam use during pregnancy (Czeizel, 1992). Clonazepam has a relatively long half-life (20-40 hours) compared to other BZDs (McElhatton, 1994), and withdrawal symptoms have been observed after exposure late in pregnancy (Fisher et al., 1985).

Clorazepate (Tranxene)

While clorazepate crosses the placenta in a limited amount, it’s metabolite nordiazepam is related to diazepam and crosses the placenta easily. Clorazepate is not teratogenic in mice, rats or rabbits, but there are no human studies available. There is a single case report of multiple malformations in an infant whose mother took clorazepate during the first trimester; the relevance of this is unknown (Patel, 1980). As such, clorazepate has an undetermined risk during pregnancy.

Diazepam (Valium)

Most reproductive studies on BZDs involve diazepam, and its reproductive risks are well reviewed (McElhatton, 1994). As previously discussed, several early case controlled studies on diazepam showed an increased risk of oral clefts, with relative risks of approximately 3-4 times the baseline risks (Aarskog, 1975; Safra and Oakley, 1975; Saxen, 1974). These early studies were criticized for their study design and other confounding factors. Studies since that time have contradicted these results, showing no increase in clefting (Rosenberg et al., 1983; Shiono and Mills, 1984; Czeizel, 1988). Prospective studies of 43 and 89 women exposed to diazepam did not show any increased risk for malformations, specifically for oral clefts (Pastuszak et al., 1994; Ornoy et al., 1998). Thus, it appears that if there is a risk of oral clefting after exposure to diazepam, it is likely to be a insignificant.

Flurazepam (Dalmane)

Human studies on flurazepam are limited to a retrospective study following 73 women exposed in the first trimester (Briggs, 1998). No increase in malformations was seen. Animal studies have not shown an increased risk for malformations (McElhatton, 1994). However, because of the paucity of information on flurazepam in pregnancy, its risks remain undetermined.

Lorazepam (Ativan)

Lorazepam crosses the placenta more slowly than diazepam, and also has a short half life (12-16 hours). Lorazepam is not considered teratogenic in mice or rats, but human reproductive data is limited. Small prospective studies (N=30; N=112) have not shown an increase or pattern to the malformations observed in women exposed to lorazepam in the first trimester (Johnson et al., 1995; Ornoy, 1998). Much of the human information has reviewed use around labor, and shows an increase in respiratory distress, decreased APGARS, problems with temperature regulation and poor feeding (McElhatton, 1994).

Oxazepam (Serax)

Oxazepam is a metabolite of diazepam. It has not been shown to increase malformations in rats, rabbits or mice (Owens et al, 1970; Miller and Becker, 1973). No malformations were noted in 89 women exposed to oxazepam in a prospective study (Ornoy, 1998). Oxazepam was, however, one of the benzodiazepines that Laegrid (1987) associated with “fetal benzodiazepine syndrome” and neonatal withdrawal.

Triazolam (Halcion)

Data on triazolam is limited to manufacturers data and a retrospective study on 138 women exposed in the first trimester; neither showed any significant increase in malformations or pattern to these malformations (Briggs, 1998), although withdrawal symptoms have been noted (Barry and St. Clair, 1987; Sakai et al. 1996).

Virtually no data is available on Halazepam (Paxipam) or Prazepam (Centrex). Animal studies on these medications do not show an increase in malformations at non-toxic levels. As such, these medications have an undetermined risk for use during pregnancy.

Summary

Use of benzodiazepines, specifically diazepam, was previously thought to be associated with an increased frequency of cleft lip and/or palate; this finding has not been supported by the majority of recent studies. Although the balance of evidence from human studies of the benzodiazepines (chiefly, diazepam) does not show first trimester usage to be teratogenic, animal studies have shown an increase in abnormal behavioral patterns after in utero exposures at levels comparable to the usual human doses. At this point, there is still no conclusive data regarding the possible behavioral teratogenicity of benzodiazepine use during pregnancy. Withdrawal symptoms can occur after fetal exposure late in pregnancy.

References:
Aarskog; Lancet 2:29, 1975.

Altshuler et al. Am J Psychiatry 153:597-8, 1996.

Anderson & McGuire. Ann Pharmacother 23:614, 1989.

Barry and St.Clair. Lancet i:1436-7, 1987.

Bergman U et al. Lancet 340:694-696, 1992.

Bracken MB. NEJM 314: 1120, 1986.

Briggs G et al. Drugs in Pregnancy & Lactation 5th ed, Williams & Wilkins, Baltimore, 1998.

Crombie DL et al. NEJM 291:1268-71, 1975.

Czeizel A. Reprod Toxicol 1(3):183, 1988.

Czeizel A et al. Eur J Epidemiol 8:122-7, 1992.

Fisher JB et al. Obstet Gynecol 66:345-55, 1985.

Gilman AG et al., eds. Goodman and Gilman’s The Pharmacological Basis of Therapeutics 8th ed., NY, Macmillan, 1990.

Godet et al.. abstract at OTIS meetings, 6/1995.

Hartz SC et al.. NEJM 292: 726-8, 1975.

Johnson K et al. Teratology 53(1):170, 1995.

Kanto JH. Drugs 23: 354-80, 1982.

Kellogg CK. Prog Brain Res 73: 207-28, 1988.

Laegrid L et al. J Peds 114:126-31, 1989.

McElhatton PR. Reprod Toxicol 8(6):461, 1994.

Milkovich et al. NEJM 291:1268-71, 1974.

Miller R, Becker B.Tox App Pharm 25:453, 1973.

Ornoy A et al. Reprod Toxicol 12(5):511-5, 1998.

Owen G et al. Tox Appl Pharm 16:556-70, 1970.

Pastuszak A et al. in Maternal-Fetal Toxicology, ed by G Koren, 2nd ed., Marcel-Decker, New York, 1994, p 77-88.

Patel DA, Patel AR. JAMA 244: 135-6, 1980.

Rosenberg L. et al. NEJM 390: 1282-5, 1983.

Rothman K et al. Am J Epidem 109:433-439, 1979.

Saxen I, Saxen L. Lancet 2:498, 1975.

Safren MJ, Oakley GP. Lancet 2:478, 1975.

Sakai T, Matsuda H, Watanabe. Eur J Pediatr 155: 1065-6, 1996.

Schardein JL (ed). Chemically Induced Birth Defects 2nd ed., Marcel-Dekker, NY; 1993. pp 209-213.

Shiono PH and Mills JM. NEJM 311: 919, 1984.

St Clair SM, Schirmer RG. ObGyn 80:843, 1992.

Zimmerman EF and Wee EL. Current Topics in Developmental Biology 19: 37-63, 1984.