PDF version

Paint and Pregnancy

This sheet talks about the risks that exposure to paint and other solvents can have during pregnancy. With each pregnancy, all women have a 3% to 5% chance of having a baby with a birth defect. This information should not take the place of medical care and advice from your health care provider.

What is paint?
Paint is pigment particles in a liquid that is called the medium. Oil paints use oil or an alkyd resin as their medium. Latex paints do not use oil and can be thinned or cleaned with water. Oil paints must be thinned or cleaned with organic solvents like paint thinner. Other ingredients are added to thin the paint, preserve it, stop rust, kill fungus, and dry the paint. Most modern household paint is latex and some is oil. Many other types of paints are used in industry, the arts and for hobbies.

Most modern paints available do not have high levels of lead or other toxic metals. However, small amounts are still present in some exterior paints. The metals are used to give the paint its color. Some older buildings still have walls painted with lead-based paint. When lead paint is sanded or disturbed it produces dust that contains lead.

Oil based paints include organic solvents. All organic solvents are liquid chemicals that may evaporate easily and can break down fats. Many solvents are made of or give off volatile organic compounds (VOCs) that contribute to indoor and outdoor air pollution. Organic solvents are widely used in industry and in residential products, often to cut grease. Exposure to these agents in the general population is unavoidable. They are used in cleaning products and cosmetics to cut grease and help the product penetrate, in aerosol sprays, in disinfectants, paints, paint strippers and thinners, varnish and many other household products.

How does exposure occur?
Paint and solvent exposure happens a lot more regularly than most people realize. You may sometimes not even realize that you are being exposed to solvents. Exposure can happen by:
• Ingestion by swallowing paint chips and dust
• Inhalation by breathing in dust and vapors
• Absorption by touching paint and other solvents

Will exposure to paint or other solvents increase the risk to my pregnancy?
There have been no consistent findings from studies of exposure to organic solvents. These studies are very limited because there is no way to measure the exact amount of exposure. Some studies have shown that solvents may affect a woman’s reproductive health by increasing the risk for preeclampsia (pregnancy induced high blood pressure) and menstrual disorders. Solvent exposure may lower fertility in both men and women. Miscarriages, low birth weight, birth defects, developmental disabilities, and higher risk of cancer for children later in life have been linked to some solvents when women were exposed to high levels or were exposed for long periods of time, for instance, in an occupational setting.

A congenital solvent syndrome from purposeful abuse has been suggested in studies that evaluated exposure from purposeful maternal abuse of organic solvents “to get high”. The reports have included prematurity, low birth weight, small head size, developmental delay, and facial features consistent with alcohol exposure. The amount of exposure in chronic abusers is typically a thousand fold times higher than in an average exposure.

At this time, there is not enough conclusive information to counsel women about exposures to organic solvents during pregnancy. If exposures are unavoidable, it is best to do as much as possible to reduce the amount of exposure to the mother, and in turn the pregnancy.

What about pregnancy and household paint use?
Currently, the assumption is that household painting involves very low levels of exposure. The recommendation is to avoid exposure to oil-based paints, leads and mercury. You should minimize exposure to latex paints that contain ethylene glycol ethers and biocides. Ideally, you should get someone else to do the job for you.

Lead based paint was commonly used prior to the 1970s, so pregnant women should avoid removing old paint because of the risk of lead exposure. According to the U.S. Food and Drug Administration, exposure to lead paint increases the likelihood of lead poisoning and mental retardation. Scraping and sanding old paint should be completely avoided. This puts higher concentrations of solvents and chemicals into the air to be inhaled. The recommendation is to have someone else do this part of the remodeling and ideally, remove yourself from the location until the project is complete. For more information about lead exposure during pregnancy, please see the OTIS lead exposure fact sheet, available at http://otispregnancy.org/pdf/Lead%208-06.pdf.

What if I have been exposed to paint already?
Currently, there are no studies that document harm to the baby during normal and incidental exposure to paint such as painting a room. The only studies that note a potential for miscarriage and malformations has to do with the higher levels of exposure through recreational use (sniffing and inhaling regularly).

If you have been exposed to paint, rest assured that the likelihood of any problems is low. According to the FDA, today’s paints do not contain lead and are probably not dangerous. Let your health care provider know of any paint exposure and together you can discuss the potential risk.

What if I was exposed to paint and I breastfeed my baby?
At this time, very little is known about exposure to paint and other solvents during breastfeeding. It is unlikely that a significant amount would enter the breast milk after a normal or incidental exposure. Whenever possible, women who are breastfeeding should try to avoid or reduce their exposure for the overall health of both the mother and the baby.

How can I reduce my risk of exposure to paint and other solvents?
Women who are pregnant or planning a pregnancy should avoid or reduce their exposure to paints and the organic solvents in many household products.

Tips to reduce exposure:

• Use products that do not need to be painted, varnished, stained or stripped.
• It is best for pregnant women not to paint or refinish. If you do paint:
  • Follow the label directions
  • Air the area out well by opening windows
  • Wear gloves and a mask with an organic vapor cartridge when possible
  • Avoid using spray paint
  • Use low-VOC latex paints.
  • Remove all paint from skin using vegetable oil or mineral oil (baby oil)
  • Do not eat or drink anything while painting.
  • Never use commercial paint thinner or stripper while pregnant
• Avoid areas being painted until the paint is dry and there is no smell.
• Choose latex paints. Use the highest quality paint you can afford. Environmentally friendly paints with low VOCs are now available.
• Other people in the household who use paints and/or organic solvents regularly should change clothes before coming home and wash their work clothes separately.
• Use fewer solvents and avoid spray products (including paints, hairspray, and oven cleaner).
• While pregnant, avoid hobbies that use solvents such as photo developing and oil painting.
• If products containing solvents are used: it is best to open windows if and if possible, wear a face mask with an organic vapor cartridge, gloves and clothing to protect your skin.

Resources
• Harms RW, et al. (eds.) 2004. Mayo Clinic Guide to a Healthy Pregnancy, Mayo Clinic, pp. 23-24.
• Heidam LZ. 1984. Spontaneous abortions among dental assistants, factory workers, painters, and gardening workers: a follow-up study. J Epidemiol Comm Health. 38:149-155.
• Hersh JH, et al. 1985. Toluene embryopathy. J Pedatr 106:922-927.
• Lindquist R, et al. 1987. Increased risk of developing acute leukemia after employment as a painter. Cancer 60:1378-1384.
• Olsen J. 1983. Risk of exposure to teratogens amongst laboratory staff and painters. Dan Med Bull 30:24-28.
• The National Institute for Occupational Safety and Health. 1999. The Effects of Workplace Hazards on Female Reproductive Health. Available from URL: http://www.cdc.gov/niosh/99-104.html.
• United States Food and Drug Administration. 1998. Dangers of Lead Still Linger. Available from URL: http://www.cfsan.fda.gov/~dms/fdalead.html.

Contributors
Kristen De Berg, BS - Genetic Counseling Intern
Christina L. Alamillo, MS, CGC - Coordinator, Illinois Teratogen Information Service

PDF Version

Second generation antihistamines and pregnancy

Volume 14, No. 1, July 2007
Inquiries to health care providers about the use of antihistamines during pregnancy are common since 20-30% of pregnant women have allergic symptoms (Ellegard et al., 2006). Avoiding known triggers (pets, mold, dust mites, and cigarette smoke), elevating the head end of the bed, and using a nasal saline spray may help avoid or alleviate symptoms (Blaiss, 2003). If these recommendations are not helpful, then providing judicious treatment with antihistamines may benefit the pregnancy by aiding in asthma control, and promoting better sleep and emotional status in pregnant women (Keles, 2004).

The first generation antihistamines (i.e. chlorpheniramine which can be found in Chlor-Trimeton Allergy) are preferred for pregnancy because they have been around longer and better studied during pregnancy. However, second generation antihistamines may be preferred by women due to their lower rates of CNS side effects, such as sedation and performance impairment (Blaiss, 2003). Therefore, this Risk Newsletter will discuss pregnancy outcomes following use of the second generation antihistamines, cetirizine (Zyrtec), fexofenadine (Allegra), and loratadine (Claritin).

Cetirizine
Cetirizine is marketed as Zyrtec. Sedative effects of this medication have been disputed; sedation seen with this medication appears to be dose dependent. While cetirizine is less sedating than the first generation antihistamines, it may be more sedating than loratadine or fexofenadine (Horak and Stubner, 1999).

The frequency of malformations was not increased among the offspring of pregnant rats or rabbits treated with up to 500 times the maximum human dose of cetirizine (Kamijima et al., 1994).

A controlled prospective study by a Canadian teratology service found no significant differences in miscarriage, birth weight, or the rates of major or minor malformations in 33 infants with first trimester cetirizine exposure (Einarson, et al., 1997). No major malformation were identified in another clinical series of 16 infants whose mothers took cetirizine during the first trimester (Wilton, et al., 1998). While reassuring, these studies are of limited value since at least 800 pregnancies would need to be evaluated to identify a 2-fold increase in congenital anomalies (Einarson, et al., 1997).

In Sweden, medicine exposures are entered into a national database following an interview with a midwife in early pregnancy. The pediatrician subsequently updates the registry with the resulting pregnancy outcomes. This dataset identified 917 women who were taking cetirizine in early pregnancy. This group did not have a significant increase in
congenital anomalies (Kallen, 2002). Although the study is large, specific timing, dose, and duration of medication is not available.

Cetirizine is also a metabolite of hydroxyzine, which has not been associated with major malformation in several clinical series totaling 187 infants who were exposed during the first trimester (Heinonen et al., 1977; Erez et al., 1971; Diav-Citrin et al., 2003; Einarson et al., 1997).

In summary, based on available pregnancy data, cetirizine is unlikely to increase the chance for congenital anomalies. Additional studies on larger numbers of pregnancies are still indicated.

Fexofenadine
Fexofenadine is marketed as Allegra. The sedative properties of fexofenadine are thought to be negligible (Horak and Stubner, 1999).

There are no published animal studies on fexofenadine. Animal studies done by the manufacturer have not shown an increase in birth defects at doses 4-30 times the human dose (Sanofi-Aventis, 2006). However, rat pups had dose related decreased weight gain and survival.

A post-marketing cohort study in England identified 23 infants whose mothers had taken fexofenadine in the first trimester. There were no major malformations although one case of “positional talipes” was reported, which is unlikely to be caused by a medicine (Craig-McFeely, et al., 2001).

Fexofenadine is a metabolite of terfenadine. Administration of terfenadine (which is no longer on the market) and fexofenadine lead to similar serum concentrations of the active metabolite so pregnancy outcomes are likely to also be similar (Loebstein et al., 1999) The Swedish Medical Birth Registry identified 1164 women treated with terfenadine in early pregnancy, and they did not have rates of congenital anomalies any higher than the general population (Kallen, 2002).

Similarly, a multicenter prospective controlled study found no congenital anomalies in 65 infants with terfenadine first trimester exposure (Loebstein et al., 1999). The mean daily dose ranged from 30-120 mg. Duration of use was not reported. While reassuring, this study has only an 80% power to rule out a 6.5-fold increased risk for malformations. Of note is that the birth weights were lower in the exposed group. However, the authors pointed out that the birth weight rates <10th percentile for gestational age did not differ between the exposed and control groups, which argues against a major clinical effect. Daily dose, duration of therapy, time of exposure, concomitant maternal asthma or steroid use did not correlate with the lower birth weight. Additional studies would need to address whether terfenadine plays a causative role in low birth weight.

While the animal and human metabolite data does not suggest an increase in birth defects, further studies on larger numbers of pregnant women are needed.

Loratadine
Loratadine is marketed as Claritin. Several studies have not seen a significant sedative effect with loratadine (Horak and Stubner, 1999). Performance impairment has been reported at doses higher than the recommended dose (Horak and Stubner, 1999).

Concern of an association between early pregnancy use of loratadine and hypospadias was raised by the Swedish Medical Birth Registry when an incidence of hypospadias twice that of the general population (but an absolute risk of <1%) was reported (Kallen, 2002). This dataset from 1995-2001 found 15 infants with hypospadias out of 2,780 loratadine exposed infants. The majority of infants had mild glandular hypospadias which makes the association with early pregnancy medicine exposure questionable in terms of biological plausibility (Diav-Citrin, et al. 2003).

Additional surveillance from 2001-2004 found only two more cases of hypospadias out of 1911 infants when 4.3 cases were expected (Kallen and Olausson, 2006). This negative follow-up study led the authors to note that the most likely explanation of the discrepancy was that the first finding was a chance event based on multiple analyses of the data.

Furthermore, two case control studies with 558 and 227 cases of hypospadias also found no association with maternal loratadine use (Werler et al., 2004; Pedersen et al., 2006). Animal data also did not support an association since McIntyre et al. (2003) reported no increase in hypospadias or other androgen mediated endpoints in animals given up to 120
times the human dose.

A prospective controlled study by the Israeli Teratogen Information Service did not find an increase in major malformations with 126 infants with exposure to loratadine during the first trimester or in the case of 175 infants with exposures at anytime during pregnancy (Diav-Citrin, et al., 2003). This study had the power to detect a 3-fold increase in major anomalies. There were also no cases of hypospadias in this study. The median daily dose was 10 mg with a median duration of 8 days.

Additionally, there was no significant differences in birth weight, premature delivery, or stillbirth. There was a significantly higher rate of miscarriage in the loratadine group, although the loratadine miscarriage rate of 11.4% was within general population rate. The authors felt a likely explanation was that the loratadine group contacted the teratogen service
significantly earlier in gestation and also had higher maternal ages compared to the control groups.

Finally, another multicenter study that also investigated the use of loratadine in the first trimester did not find an increase in congenital anomalies in 143 infants (Moretti et al., 2003). This study had an 80% power to detect a 3.5- fold increase in malformations. No cases of hypospadias were observed in the exposed group either. There was also no significant increase in miscarriage, birth weight, or gestational age at delivery. The median daily dose was 10 mg and median total dose in the first trimester was 50 mg (range of 10-1470 mg).

Based on the combined animal and human studies, loratadine does not appear to significantly increase the risk of hypospadias or non-genital congenital anomalies.

Pseudoephedrine
All of these second generation antihistamines can be purchased in formulation with pseudoephedrine, i.e. Zyrtec-D, Allegra-D, and Claritin-D. Studies have not associated pseudoephedrine with congenital anomalies in over 2000 pregnanicies (Jick et al., 1981; Aselton et al., 1985; Heinonen et al., 1977; Rosa, 1993).

However, weak associations with vascular disruptions, such as gastroschisis, have been reported, and cigarette smoking may further exacerbate the risks. (Werler et al., 1992, 2002, 2004). Although weak, these associations have not been entirely discounted due to the vasoactive properties of decongestants. If truly causative, absolute risks would still be <1%. Preferentially, antihistamines are taken alone, especially during the first trimester.

Summary
The use of first generation antihistamines is preferred in pregnancy over the use of second generation antihistamines due to their longer availability on the market. However, women who do not respond to chlorpheniramine can be counseled that the available studies for cetirizine, fexofenadine, and loratadine do not show an increased risk for congenital anomalies. Larger numbers of human pregnancies are still needed, particularly for fexofenadine. Avoidance of combination therapy with pseudoephedrine, at least in the first trimester, is also preferred.

Contributors
Kate Durda, BA
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

References
· Aselton P et al. (1985): First-trimester drug use and congenital disorders. Obstet Gynecol 65(4):451-455.
· Blaiss MS (2003). Management of rhinitis and asthma in pregnancy. Ann Allergy Asthma Immunol 90:16-22.
· Craig-McFeely PM et al. (2001). Evaluation of the safety of fexofenadine from experience gained in general practice use in England in 1997. Eur J Clin Pharmacol 57:313-320.
· Diav-Citrin O et al. (2003). Pregnancy outcome after gestational exposure to loratadine or antihistamines: a prospective controlled cohort study. J Allergy Clin Immunol 111:1239-1243.
· Einarson A et al. (1997). Prospective controlled study of hydroxyzine and cetirizine in pregnancy. Ann Allergy Asthma Immunol 78:183-186.
· Ellegard E (2006) Pregnancy Rhinitis. Immunol and Allergy Clin of North Am 26:119-135.
· Erez S et al. (1971) Double-blind evaluation of hydroxyzine as an
· antiemetic in pregnancy. J Reprod Med 7(1):57-59.
· Heinonen OP et al. (1977) Birth Defects and Drugs in Pregnancy. Littleton, Mass.: John Wright-PSG, 1977, pp 335-337, 346-347, 438-439.
· Horak F and Stubner UP (1999). Comparative tolerability of second generation antihistamines. Drug Saf 20:385-401.
· Jick H et al. (1981) First-trimester drug use and congenital disorders. JAMA 246(4):343-346.
· Kallen B (2002). Use of antihistamine drugs in early pregnancy and delivery outcome. J Matern Fetal Neonatal Med 11:146-152.
· Kallen B and Olausson PO (2001). Monitoring of maternal drug use and infant congenital malformations. Does loratadine cause hypospadias? Int J of Risk & Safety in Med 14:115-119.
· Kallen B and Olausson PO (2006). No increased risk of infant hypospadias after maternal use of loratadine in early pregnancy. Int J Med Sci 3:106-107.
· Kamijima M et al. (1994) [Reproductive and developmental toxicity studies of cetirizine in rats and rabbits.] Kiso to Rinsho 28(7):1877-1903.
· Keles N (2004) Treatment of allergic rhinitis during pregnancy. Am J Rhinol 18:23-28.
· Loebstein R et al. (1999) Pregnancy outcome after gestational exposure to terfenadine: A multicenter, prospective controlled study. J Allery Clin Immunol 104:953-6.
· McIntyre BS et al. (2003). Effects of perinatal loratadine exposure on male rat reproductive organ development. Reprod Toxicol 17:691-697.
· Moretti ME et al. (2003). Fetal safety of loratadine use in the first trimester of pregnancy: A multicenter study. J Allergy Clin Immunol 111:479-483.
· Pederson L et al. (2006) Maternal use of loratadine during pregnancy and risk of hypospadias in offspring. Int J Med Sci 3(1):21-25.
· Rosa F (1993): Personal Communication. Cited in: Briggs et al. Drugs in Pregnancy and Lactation: A Reference Guide to Fetal and Neonatal Risk, 7th ed. Philadelphia, Pa.: Lippincott Williams & Wilkins, 2005, p 1371.
· Sanofi-Aventis (2006). Allegra Product Information.
· Werler MM et al. (1992) First trimester maternal medication use in relation to gastroschisis. Teratol 45:361-367.
· Werler MM et al. (2002) Maternal medication use and risks of gastroschisis and small intestinal atresia. Am J Epidemiol 155(1):26-31.
· Werler MM et al. (2004). Evaluation of an association between loratadine and hypospadias - United States, 1997-2001. MMWR 53:219-221.
· Werler MM et al. (2004) Vasoactive exposures, vascular events, and hemifacial microsomia. Birth Defects Res A Clin Mol Teratol 70(6):389-395.
· Wilton LV et al. (1998). The outcomes of pregnancy in women exposed to newly marketed drugs in general practice in England. Br J Obstet Gynaecol 105:882-889.

PDF Version

Neonatal Complications Associated with Late Pregnancy Use of Selective Serotonin Reuptake Inhibitors

Volume 13, No. 2, April 2006

The prevalence of depression among women peaks during the childbearing years (Burke et al., 1991). With up to 14% of pregnant women displaying symptoms of depression, treatment of major depression during pregnancy is quickly becoming a major public concern (Evans et al., 2001).

In addition to concerns that women with untreated depression are less likely to follow through with prenatal care and more likely to develop unhealthy behaviors or consider suicide, some studies have suggested that maternal depression is associated with pregnancy complications such as pre-term birth and low birth weight, as well as adverse  cognitive/emotional effects in the child (reviewed in Bonari et al., 2004; Mian 2005). Further, for pregnant women with a history of major depression, relapse occurred during pregnancy in 68% of those who discontinued their medications, as compared to 26% of those that continued taking medication for their depression (Cohen et al., 2006). The potential for relapse and the possible adverse effects of depression itself on the pregnancy need to be considered when discussing whether or not a specific patient should remain on antidepressants throughout pregnancy.

Selective serotonin reuptake inhibitors (SSRIs), a class of antidepressants that increase the levels of circulating serotonin in the body, are just as effective as older tricyclic antidepressants and have fewer side effects (nausea, insomnia, sexual dysfunction, etc.) and higher tolerability in general (reviewed in Zohar and Westenberg, 2000). As a result, the prevalence of those taking SSRIs to treat their depression has increased dramatically in the past decade (Meijer et al., 2004).

A number of studies have found no association with congenital anomalies and SSRIs with use during the first trimester of pregnancy (reviewed in Nonacs and Cohen, 2003; Wen and Walker 2004). In contrast to this reassuring data, a retrospective and unpublished investigation by GlaxoSmithKline, the makers of paroxetine (Paxil), suggests an increased risk for cardiovascular defects (most commonly ventricular septal defects) of 2% compared to 1% in the general population(reviewed in Williams and Woolerton 2005; GlaxoSmithKline study EPIP083 2005). Additional studies are still needed for confirmation since controlled prospective studies have not seen an increase in congenital anomalies with paroxetine.

There is also accumulating reports regarding concerns of SSRI use after the first trimester. Therefore, this newsletter will focus on the neonatal effects that have been associated with use of SSRIs during the second half of pregnancy, such as poor neonatal adaptation, and the recent reported association with persistent pulmonary hypertension of the newborn (PPHN).

Poor Neonatal Adaptation

Goldstein (1995) reviewed 112 prospective pregnancy outcomes on the use of fluoxetine (Prozac) that were voluntarily reported to the manufacturer, Eli Lilly. He found that 15/115 (13%) of infants experienced different postnatal complications, such as irritability, hyperbilirubinemia, or sleepiness, which he noted was similar to the National Hospital Discharge Survey. There was also no pattern of complications and no dose relationship to further a cause and effect relationship between fluoxetine and the complications. However, absence of a specific control group and a high rate of unknown outcomes limited his conclusions.

Subsequently, Chambers et al. (1996) prospectively followed 73 pregnant women taking fluoxetine during the third trimester. The authors reported poor neonatal adaptation among 31.5% of infants exposed to fluoxetine in the third trimester, compared to 8.9% among infants exposed early in pregnancy (Chambers et al. 1996). With exclusion of the premature infants, 23% of late exposed infants were admitted to the special care nursery compared to 9.5% of early exposed infants or 6.3% of infants never exposed to fluoxetine. Health care workers, however, were not blinded to maternal medication use when examining the infants, a possible study bias.

Similar results were noted in a study of 55 infants exposed to paroxetine during the third trimester. In this study, 22% of infants experienced neonatal complications (including 3 premature infants), as compared to 6% of unexposed infants or infants exposed only in the first and second trimester (Costei et al., 2002). Of the 12 infants that had complications, respiratory distress was noted in nine, hypoglycemia in two, and jaundice in one. Symptoms disappeared in one to two weeks.

Oberland et al. (2004) performed a prospective cohort on three small groups of infants. Twenty-eight infants were exposed to SSRIs alone (paroxetine, fluoxetine, and sertraline), 18 infants had exposure to an SSRI and the benzodiazepine clonazepam, and 23 control infants had no medication or depression exposure. Twenty-five percent of the SSRI group alone, 39% of the SSRI group plus clonazepam, and 9% of control infants had symptoms of poor neonatal adaptation.

Symptoms included mild respiratory distress, transient tachypnea of the newborn, and hyptotonia. All symptoms resolved in 48 hours and no significant differences in development were noted at follow-up at 2 and 8 months of age using the Bayley Scales of Infant Development.

Based on their findings, the authors recommended avoiding polypharmacy when possible and specifically monitoring mothers and infants when paroxetine was used in combination with clonazepam. They also noted that their data should not preclude the urgency to treat maternal depression or anxiety when warranted.

Sivojelezova et al. (2005) prospectively followed 132 women taking citalopram (Celexa) in whom 54% of the total group took citalopram throughout pregnancy. Following third trimester exposure, 16% of infants were admitted to the special care nursery versus 4% of infants not exposed to citalopram during the third trimester. Additionally, Malm et al. (2005) used Finnish population based registries to identify 597 women who purchased SSRIs in the third trimester. Citalopram(N=228) and fluoxetine (N=239) were most commonly purchased while paroxetine (N=64), sertraline (N=41), and fluvoxamine (N=27) were also purchased. There was a small but significant difference in that 15.4% were treated in the special care nursery versus 11.2% with only first trimester purchase. An obvious criticism of this study is that purchasing the medicine does not necessarily mean the medicine was taken during pregnancy.

Levinson-Castiel (2006) identified 60 infants with prolonged exposure to SSRI paroxetine (N=37), fluoxetine (N=12), citalopram (N=8), venlafaxine (n=2), and sertraline (N=1). Symptoms of neonatal abstinence syndrome were present in 30% of the exposed infants versus none of the control infants. Clinical assessment was not made blinded and medicine use was based on maternal report. The most common symptoms were tremors, gastrointestinal disturbances (poor feeding, vomiting, loose stools), and sleep disturbances. Maximum mean daily symptoms occurred within the first 48 hours. None of the infants in this study required any treatment for the symptoms.

A dose relationship with increasing symptoms and increasing dose was seen for paroxetine (the only medicine that could be studied for dosing) but no cut-off point of an increased risk could be identified due to the small sample size. Based on their findings, the authors recommended that exposed infants should be closely monitored by using a standardized protocol for a minimum of 48 hours and should not be discharged early from the hospital.

The results of a meta-analysis of studies involving late gestational SSRI exposure, including many of the ones described previously, indicated an overall neonatal behavioral syndrome risk ratio of 3.0 (95% CI, 2.0-4.4) (Moses-Kolko E 2005). Paroxetine and fluoxetine were reported most often but may reflect their more common use. The authors noted that tapering and then discontinuing SSRIs a couple weeks prior to the due date and then resuming treatment right after delivery may be an option for some women. They did note that a late tapering has not been proven effective for avoidance of neonatal complications and that it puts patients at risk for postpartum depression.

It is unclear whether the symptoms noted in some infants born after in utero exposure to SSRIs is due to withdrawal from medication at delivery or to serotonergic overstimulation due to exposure late in pregnancy. The symptoms of these syndromes overlap in adults, and appear to overlap in infants as well (Jaiswal et al., 2003).

Persistent Pulmonary Hypertension of the Newborn

Persistent pulmonary hypertension of the newborn (PPHN) is described as a failure of normal pulmonary vascular relaxation shortly after birth, which ultimately results in unoxygenated blood being shunted into the systemic circulation (via patent ductus arteriosus and/or foramen ovale) and profound hypoxemia (reviewed in Dakshinamurti 2005). This condition is associated with substantial morbidity and, despite treatment, can result in death.

A recent study published in the New England Journal of Medicine found that SSRI use after the 20th week of gestation is associated with PPHN (Chambers et al., 2006). This study was prompted by the authors original cohort study that identified two infants with PPHN following third trimester fluoxetine use. For this case-control study of 377 infants with PPHN and 836 controls, mothers were interviewed within 6 months of delivery regarding their medical and obstetric histories, habits, occupations, and medication use during the period of 2 months before conception and the end of pregnancy. The mothers were explicitly asked whether they had taken medications for depression, and if they responded positively, they were provided a list of antidepressant medications from which to identify the one(s) they had taken. The specific SSRI medications that participants reported included citalopram, fluoxetine, paroxetine, and sertraline. Doses of medication and the number of women reporting use of each medicine was not listed.

Fourteen infants with PPHN had been exposed to SSRIs, as compared to only 6 control infants. This resulted in an odds ratio of 6.1. Given the general population rate of 1-2/1000 for PPHN, this translates into an absolute risk of 0.6-1.2%. 12/14 mothers with infants with PPHN continued the SSRI at least into the eights months of pregnancy. Using 26 gestational weeks as a cutoff therefore yielded identical results.

The authors of this study point out that such an association cannot establish causality, but that there is biological plausibility to their finding. The lungs have been reported to act as a reservoir for SSRIs, the vasoconstrictive properties of SSRIs may increase pulmonary vascular resistance, and the mitosis-inducing properties of the drugs may result in the over-proliferation of the smooth muscle cells in the lung (Chambers et al., 2006). Further, the authors also suggest the idea that SSRIs affect the synthesis of nitric oxide, which is known to play a role in regulation of vascular tone and reactivity as yet another mechanism for the occurrence PPHN in SSRI-exposed infants (Chambers et al., 2006). They propose that PPHN may represent the severe end of the neonatal complications (Chamber et al., 2006).

Although the study design was the most appropriate in order to assess this rare outcome, its retrospective design makes it vulnerable to recall bias, especially with prompting women with medicine names. The authors point out that the same association was not found with tricyclic antidepressants even with medicine name prompting. It is still important to replicate these findings before definitive conclusions can be drawn.

Summary

Use of SSRIs late in pregnancy is associated with an approximate 20-30% chance for nonspecific and typically transient neonatal complications. Therefore, these infants should be carefully monitored in the newborn period. While an initial study on these affected infants did not find development impairment at eight months of age, longer term neurobehavioral evaluations in this subpopulation are still needed. It should be noted that the SSRI associated neonatal complications can also be seen with other types of antidepressants and in a smaller number of infants whose mothers do not take any antidepressants.

An approximate 1% link between PPHN and late pregnancy SSRI needs to be replicated for verification but may represent the severe end of these neonatal complications. These findings act as a reminder that a woman’s individual clinical state should be considered when deciding to keep a patient on a psychotropic medicine throughout pregnancy. Due to concerns with untreated maternal depression, continuing treatment in women with clinical depression may still be the least risky option for many woman.

References

• Bonari L, et al.(2004). Perinatal risks of untreated depression during pregnancy. Can J Psychiatry 49(11):726-734.
• Burke KC, et al. (1991). Comparing age at onset of major depression and other psychiatric disorders by birth cohorts in five US community populations. Arch Gen Psychiatry 48(9):789-95.
• Chambers C, et al . (1996). Birth outcomes in pregnant women taking fluoxetine. New Engl J Med 335:1010-5.
• Chambers C, et al. (2006). Selective serotonin-reuptake inhibitors and risk of persistent pulmonary hypertension of the newborn. New Engl J Med 354:579-87.
• Cohen L, et al. (2006). Relapse of major depression during pregnancy in women who maintain or discontinue antidepressant treatment. JAMA 295(5):499-507.
• Costei A, et al. (2002). Perinatal outcome following third trimester exposure to paroxetine. Arch Pediatr Adolesc Med 156:1129-32.
• Dakshinamurti S (2005). Pathophysiologic mechanisms of persistent pulmonary hypertension of the newborn. Pediatr Pulmonol 39(6):492-503.
• Evans J, et al. (2001). Cohort study of depressed mood during pregnancy and after childbirth. BMJ 323(7307):257-60.
• GlaxoSmithKline study EPIP083 (2005). GSK medicine: bupropion and paroxetine. Epidemiology study: preliminary report on bupropion in pregnancy and the occurrence of cardiovascular and major congenital malformation. Available: http://ctr.gsk.co.uk/summary/paroxetine/epip083.pdf (Accessed 30 March 2006).
• Goldstein D (1995). Effects of third trimester fluoxetine exposure on the newborn. J Clin
• Psychopharmacol 15:417-20.
• Jaiswal S, Coombs R, Isbister G (2003). Paroxetine withdrawal in a neonate with historical and
• laboratory confirmation. Eur J Pediatr 162:723-4.
• Levinson-Castiel R, at al. (2006) Neonatal abstinence syndrome after in utero exposure to selective serotonin reuptake inhibitors in term infants. Arch Pediatr Adolesc Med 160:173-6.
• Malm H, et al. (2005). Risks associated with selective serotonin reuptake inhibitors in pregnancy. Obstet Gynecol 106:1289-96.
• Meijer W, et al. (2004). Incidence and determinants of long-term use of antidepressants. Eur J Clin Pharmacol 60(1):57-61.
• Mian A (2005). Depression in pregnancy and the postpartum period: balancing adverse effects
• of untreated illness with treatment risks. J Psychiatr Prac 11:389-396.
• Mose-Kolko EL, et al. (2005) Neonatal signs after late in utero exposure to serotonin reuptake inhibitors. JAMA 293:2372-2383.
• Nonacs R, Cohen L (2003). Assessment and treatment of depression during pregnancy: an
• update. Psychiatr Clin North Am 26(3):547-62.
• Oberland TF, et al. (2004) Pharmacologic factors associated with transient neonatal symptoms following prenatal psychotropic medication exposure. J Clin Psychiatry 65:230-237.
• Sivojelezova A, et al. (2005) Citalopram use in pregnancy: Prospective comparative evaluation of pregnancy and fetal outcome. Am J Obstet & Gynecol 193:2004-9.
• Wen S, Walker M (2004). The use of selective serotonin reuptake inhibitors in pregnancy. J Obstet Gynaecol Can 26(9):819-22.
• Williams M, Wooltorton E (2005). Paroxetine (Paxil) and congenital malformations. CMAJ 173(11):1320-1.
• Zeskind P, Stephens L (2004). Maternal selective serotonin reuptake inhibitor use during
• pregnancy and newborn neurobehavior. Pediatrics 113:368-75.
• Zohar J, Westenberg H (2000). Anxiety disorders: a review of tricyclic antidepressants and selective serotonin reuptake inhibitors. Acta Psychiatr Scand Suppl 403:39-49.

Contributors

Erin Rooney, BA
Genetic Counseling Intern

Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service

Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

Organic Solvents During Pregnancy: An Update on Occupational Exposure

Volume 13, No. 1, January 2006

Organic solvents are a diverse group of liquids that are used to dissolve other materials like oils, resins, and rubber (McMartin et al., 1998). Examples of organic solvents include: aliphatic hydrocarbons (mineral spirits, varnish, kerosene); aromatic hydrocarbons (benzene, toluene, xylene); halogenated hydrocarbons (carbon tetrachloride, trichloroethylene); aliphatic alcohols (methanol); glycols (ethylene glycol); and glycol ethers (methoxyethanol) (McMartin et al., 1998). While it is likely erroneous to make general statements about organic solvents due to their diverse nature, human reproductive studies have only studied organic solvents collectively. This newsletter will provide an updated review of these studies. See 1995 Risk Newsletter Volume 3(3) for the original newsletter.

Exposure to organic solvents occurs most frequently from inhalation or skin contact. Exposure is nearly unavoidable in the general population since organic solvents are present in many household products such as paints, aerosol sprays, spot removers, lighter fluids, inks, and gasoline (Koren 2005). Household exposure is typically of a low level, episodic nature. This type of exposure is likely to be of little concern although conservative measures should include minimizing use, wearing protective clothing, such as gloves, making sure ventilation is adequate, or ideally having someone else use the product in the woman’s absence. Occupational exposure is more of a concern because it typically involves higher dose, chronic exposure. Organic solvents are used in a wide range of industries such as dry cleaners, laboratories, paint manufacturers, chemical manufacturers of inks and plastics, beauty salons, carpentry, and funeral services (Koren, 2005). Moreover, some of these fields are female dominated.

Intentional Solvent Abuse
Individuals may purposefully inhale organic solvent vapors to attain a feeling of euphoria by coating rags with spray paint and directly inhaling the rag as it covers their nose and mouth. Chronic toluene abuse (one-four 16 ounce cans per day) is associated with a myriad of toxic symptoms such muscle weakness, neuropsychiatric disturbances, and renal tubular acidosis. (Pearson et al., 1994). Chronic abusers typically reach levels 50 x the Occupational Safety and Health Administration (OSHA) permissible levels(Wilkins-Haug, 1997).

Although causation is not proven, a congenital solvent syndrome from purposeful abuse of toluene or gasoline has been suggested by case series and uncontrolled cohorts. These reports have included premature delivery, low birth weight, microcephaly, develop-mental delay, growth retardation, and characteristic facial features similar to infants with fetal alcohol syndrome (Arnold et al., 1994; Pearson et al., 1994). While fetal alcohol syndrome was not ruled out, case reports have also described infants with renal tubular dysfunction which is associated with solvent and not alcohol abuse (Lindermann, 1991; Erramouspe et al., 1996). Additionally, some infants have smelled of solvents at birth (Koren, 2005).

Occupational Exposure Occupational exposures are difficult to assess for several reasons. First epidemiological reproductive data is typically sparse and high dose animal data is difficult to interpret to lower level human exposure. Secondly, the dose of exposure is also difficult to gauge unless specific airborne studies have been performed by an industrial hygienist. Odor is often an inaccurate way to assess the dosage of most organic solvents (McMartin et al, 1998).

Infrequently, organic solvents (such as toluene) can be monitored via blood or urine tests. Thirdly, occupational exposures typically include multiple exposures which confound outcome information for any one agent. In summary, available studies are limited by the lack of quantified measure of the exposure and the fact that exposures take place in diverse settings with different doses, durations, and combinations of chemicals.

Miscarriage and Major Malformations
Occupational studies have yielded conflicting results on whether rates of miscarriage and congenital anomalies are increased following maternal exposure. McMartin et al. (1998) performed a meta-analysis of epidemiological studies on pregnancy outcomes following maternal organic solvent exposure. Five retrospective studies were included for spontaneous abortion (N= 2,899 patients) and five retrospective studies were utilized for major malformation analysis (N= 7,036 patients). While some prior studies have suggested an increased rate of miscarriage, the rate of spontaneous abortion was not significantly increased when analyzed in this meta-analysis. However, the rate of major malformations was significantly increased with an odds ratio of 1.64 (CI 1.16-2.30). Assuming a background risk of 3%, this study suggests the absolute risk for malformations would be 4.9%.

Limitations of the McMartin et al. (1998) meta-analysis include that the studies analyzed were looking at different organic solvents, unknown dosages, and wide durations of exposure in a wide range of occupational environments. Despite these limitations, the authors supported the current recommendation that pregnant women should limit their exposure to organic solvents as much as possible. The authors also pointed out that these risks should be further investigated in a prospective study. Khattak S et al. (1999) prospectively followed 125 pregnant women who were occupationally exposed to organic solvents. All women worked for at least the entire first trimester.

Hours were not documented. The majority of women were factory workers, laboratory technicians, artists, chemists, painters, and printers. The most commonly reported organic solvent exposures were aliphatic and aromatic hydrocarbons, phenol, trichloroethylene, vinyl chloride, and acetone. The exposed group had a significantly higher rate of malformations compared to the control group. There were 13 major malformations in the exposed group compared to only one malformation in the control group. While the control group had an unusually low rate of malformations, the exposed rate was still greater than historic controls. Additionally, a dose response relationship was suggested in that at least 12/13 of the infants with malformations occurred in women who reported work-related illness such as irritation of the eyes or respiratory system, headaches, and breathing difficulties.

There was no pattern to the malformations to suggest cause and effect. Malformations included ventricular septal defect, clubfoot, laryngolmalacia, diaphragmatic hernia, neural tube defect, congenital deafness, micropenis, cloacal extrophy, left inguinal hernia, congenital hydronephrosis, neuronal migration defect, and hemivertebrae. However, the authors argued that diverse exposures should not be expected to create a homogenous pattern. The authors concluded that additional studies are needed to confirm their findings but in the interim it is prudent to minimize women’s exposure to organic solvents and particularly to take precaution to avoid symptomatic exposure.

Neurological effects
Organic solvents like toluene are considered neurotoxins since acute exposures have documented effects on the central nervous system in adult workers. A subset of adults with a long history of occupational exposure have cognitive deficits and women with purposeful abuse have delivered infants with developmental delay (Koren, 2005). Therefore, fetal exposure raises a significant concern for adverse cognitive function following exposure to known neurotoxins. Similar to major malformations and miscarriage, results however have been inconsistent.

Eskenzai et al. (1988) assessed neurocognitive functioning and growth in 41 children age 3-4 years old whose mothers had been occupationally exposed to organic solvents. Using maternal reports and the McCarthy Scales of General Abilities, no differences in neurobehavioral development or growth were identified in the exposed versus control group.

Till et al. (2001a) also evaluated long-term cognitive and behavioral functioning of 28 children ages 3-7 years old who were exposed to a wide range of organic solvents in utero. No group differences were observed on measures of attention, visuo-spatial ability, or fine-motor ability. However, exposed children scored significantly lower on receptive language, expressive language, and graphomotor ability, suggesting detrimental effects on selective cognitive functions.

Laslo-Baker et al. (2004) examined the long-term neurodevelopment effects of 32 children age 3-9 years old whose mothers had been exposed to organic solvents at work during their pregnancy. Mothers reported exposures to a total of 78 different organic solvents for 1-40 hours per week (mean, 24 hours) and for 8-40 weeks (mean, 32 weeks) during their pregnancy. Exposed mothers reported a high level of protective equipment. Occupations listed were diverse including painter, laboratory technician, factory worker, hair stylist, graphic designer, funeral embalmer, and science teacher. After controlling for demographic variables and maternal IQ and education, there were no significant differences in global, verbal, or performance IQ. The authors however reported differences in more subtle areas of neurodevelopment. Exposed children had lower test scores in subtests of recall, attention, and language. No dose relationship (which would further causation) was found between the length and total hours of exposure and any of neurodevelopmental scores.

Visual impairments
Noting that impairment of color vision discrimination can be altered by certain solvent exposures in adult workers, Till et al. (2001b, 2005) conducted two studies to examine visual impairments in infants and children whose maternal parents were occupationally exposed to organic solvents during pregnancy.

Till et al. (2001b) measured color vision and visual acuity in 32 exposed children whose mothers worked in diverse occupations with multiple exposures during their pregnancies. Overall, results showed that exposed children had significantly higher error scores on color discrimination and visual acuity. Three of the 32 exposed children suffered from clinical red-green color vision loss compared to none of the control children. The authors commented that these visual deficits are of concern because they may have implications for higher level cognitive functioning, such as learning to read. A dose response relationship, which would argue for causation, was not found.

In 2005, Till et al. tested 21 exposed infants in which maternal qualitative exposure levels were gathered prospectively. This study found a significant decrease in contrast sensitivity as well as grating acuity (dependant upon the level of exposure) in exposed infants compared to controls. Regarding color vision, 26.3% of exposed infants showed abnormal red-green color vision compared to 0% of controls. These findings suggest that prenatal solvent exposure is associated with selective visual deficits.

Summary
Maternal intentional abuse of organic solvents has suggested that a congenital solvent syndrome, similar to fetal alcohol syndrome, may exist. Effects of doses not toxic to the mother warrant further investigation. Available occupational exposure studies are severely limited in that they study multiple diverse exposures together and that there is no quantified measure of the exposures.

Not surprising, the available data is inconsistent. There is some evidence of an increased risk for congenital anomalies, particularly in women who reported symptoms related to their occupational exposure. Initial data suggests that there may be detrimental effects on subtle cognitive functions or visual color acuity. Given the inconclusive but concerning information, occupational exposure to known neurotoxins should be minimized or ideally avoided. If air monitoring is available, OSHA dose standards that are established for adult health, and may or may not adequately protect fetal health, should not be exceeded. Mothers are encouraged to limit the time they are exposed to organic solvents and to wear protective clothing, such as solvent resistant gloves and splash-proof eye goggles. Respirators need to be specifically approved for organic solvents and individuals would need to be assessed and fitted to use them. Engineering controls, such as a chemical hood, which would contain chemicals in their own ventilation system are preferred.

References
· Arnold GL, Kirby RS, Langendoerfer S, Wilkins-Haug L (1994). Toluene embryopathy: clinical delineation and developmental follow-up. Pediatrics 93(2):216-20.
· Erramouspe J, Balvez R, Fischel D (1996) Newborn renal tubular acidosis associated with prenatal maternal toluene sniffing. J Psych Drugs 28:201-4.
· Eskenazi B, Gaylord L, Bracken MB, Brown D (1988) In utero exposure to organic solvents and human neurodevelopment. Dev Med Child Neurol 30(4):492-501.
· Khattak S, K-Moghtader G, McMartin K, Barrera M, Kennedy D, Koren G (1999) Pregnancy outcome following gestational exposure to organic solvents: a prospective controlled study. JAMA 281:1106-1109.
· Koren G (2005) Occupational toxicology of organic solvents-the reproductive context. 18th International Conference of Organization of Teratology Information Services.
· Laslo-Baker D, Barrera M, Knittel-Keren D, Kozer E, Wolpin J, Khattak S, Hackman R, Rovet J, Koren G (2004) Child neurodevelopment outcome and maternal occupational exposure to solvents. Arch Pediatr Adolesc 128:956-961.
· Lindermann R (1991) Congenital renal tubular dysfunction associated with maternal sniffing of organic solvents. Acta Paediatr Scand 80(8-9):882-4.
· McMartin KI, Chu M, Kopecky E, Einarson TR, Koren G (1998) Pregnancy outcome following maternal organic solvent exposure: a meta-analysis of epidemiologic studies. Amer J Indust Med, 34:228-292.
· Pearson MA, Hoyme HE, Seaver LH, Rimsza ME (1994) Toluene embryopathy: delineation of the phenotype and comparison with fetal alcohol syndrome. Pediatrics 93(2):211-215.
· Till C, Koren G, Rovet JF (2001) Prenatal exposure to organic solvents and child neurobehavioral performance. Neurotoxicol Teratol 23(3):235-45.
· Till C, Westall CA, Rovet JF, Koren G (2001) Effects of maternal occupational exposure to organic solvents on offspring visual functioning: a prospective controlled study. Teratology 64(3):134-41.
· Till C, Westall CA, Koren G, Nuiman I, Rovet JF (2005) Vision abnormalities in young children exposed prenatally to organic solvents. Neurotoxicology 26(4):599-613.
· Wilkins-Haug L (1997) Teratogen Update:Toluene. Teratology 55:145-151.
· Wilkins-Haug L and Gabow PA (1991) Toluene abuse during pregnancy: obstetric complications and perinatal outcomes. Obstet Gynecol 77:504-9.

Contributors
Lauren Bowling, BS
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

Paternal exposure to chemotherapy

Volume 12, No. 5, September 2005

How many environmental agents with mutagenic properties have been found to increase the rate of genetic disorders in humans (Brent, 2005)? The answer: None. Infertility is the only established outcome following certain paternal exposures. For example, clinicians were never able to document an increase of genetic disease in the offspring of male atomic bomb survivors. Unfortunately, the topic of male-mediated teratogenesis has not been thoroughly studied and the available  research is inadequate to rule out any increased risk for adverse pregnancy outcome. Because an increase in genetic disease would be at a much lower magnitude than an increase in congenital anomalies following a maternal teratogen exposure, further studies with very large populations are still needed.

The maturation of sperm takes approximately 3 months. Mutations are more likely to occur in the actively reproducing cells like spermagonia and are less likely to occur in female oocytes (Davis et al., 1992). Therefore, preconception exposures are a greater theoretical concern for males. There are several mechanisms by which paternal exposure could theoretically result in reduced fertility or exert teratogenic effects on the fetus:
·Medicines could exert a direct effect on sperm causing anatomic abnormalities with reduced sperm motility or function or even induce chromosomal anomalies or new gene mutations
·Medicines could interfere with the pituitary-hypothalmic function (decrease testosterone or libido) and subsequently reduce fertility
·Medicines could lead to direct exposure of the developing fetus through contaminated seminal fluid via unprotected intercourse

This newsletter will focus on studies addressing the first mechanism. Since ionizing radiation and antineoplastic drugs are biologically plausible agents for inducing mutations in sperm, possible paternal teratogenecity with various cancer treatments will be discussed.

Survivors of childhood cancer
There is concern for long lasting effects in the sperm of male survivors since cancer therapies are targeted to induce cell damage. These concerns are further prompted by the presence of higher rates of chromosomal abnormalities in sperm of some successfully treated patients (Martin et al., 1999; Frias et al., 2003).

However, while male infertility has been documented with germ cell loss due to radiation or antineoplastic drug treatment, an increased frequency of adverse pregnancy outcomes has not been documented in the offspring of survivors of childhood cancer.

For example, the National Cancer Institute collaborated with three hospital-based cancer registries and two population based registries to evaluate childhood cancer (diagnosis < age 20 years) survivors’ pregnancy outcomes. There were no differences in the rates of cytogenetic diseases, singlegene defects, or simple malformations in the offspring when compared with sibling controls (Byrne et al., 1998). Genetic disease occurred in 3.4% of 2,198 offspring of survivors, compared with 3.1% of 4,544 offspring of controls (P = 0.3; not significant).

Several other large population studies report the same findings. This includes the Childhood Cancer Survivor Study, a multi-institutional retrospective cohort study started in 1994. This database was used to review 4,214 livebirths from childhood cancer survivors. Again there were no significant differences between the two study groups. The total rate of cytogenetic abnormalities, single gene defects, and malformations was 3.7% in the survivor offspring and 4.1% in the siblings (Boice et al., 2003). Additionally, when cancers with a known single gene inheritance were excluded, no increase
in offspring childhood cancer was seen.

A study conducted using the Danish Cancer Registry of childhood cancer survivors compared the pregnancy outcomes to their unaffected siblings (Winther et al., 2004). The cancer survivor group had 2,630 live births from 4,676 childhood cancer survivors. There was no difference in the rate of chromosomal abnormalities between the groups.

It should be noted that a limitation of these studies is that the individual treatment regiments have not been evaluated separately. There may be subtle long-term differences between treatment types that are not detectable by grouping all survivors together(Wyrobek et al., 2005).

Testicular Cancer
Testicular cancer and Hodgkin’s disease represent two of the most common cancers in young males.

Several researchers have noted oligospermia in patients with testicular cancer even prior to treatment (Gandini et al., 2003). Bahadur et al. (2005) analyzed semen quality in 314 males before and after gonadatoxic therapy over a 26 year period. The testicular cancer group (N=102) had the lowest level of sperm concentration of all disease categories before treatment. Following cytotoxic treatment, this group had the lowest level of azoospermia (12%) but the highest level of oligospermia (38%). The rates of normal sperm counts after treatment (50%), however, were also the highest of all the cancer types evaluated.

Testicular cancer is commonly treated with the polychemotherapy regimen bleomycin, etoposide and cisplatin (BEP). Recovery of sperm function following treatment has not been predictable. While azoospermia following BEP occurred initially in all patients in the Bahadur et al. study, 50-80% of men recovered some level of motile sperm 2-5 years following treatment.

Petersen et al. (1994) found a dose-dependent effect of cisplatin and impaired spermatogenesis by analyzing sperm samples in 33 patients treated with a conventional dose of BEP and 21 patients treated with high dose BEP. The conventionally treated group had a higher sperm count (19% azoospermic) than the high dose group (47% azoospermic) following treatment.

Chromosomal abnormalities in sperm were found to be increased up to one year post treatment for testicular cancer (Martin et al., 1999) but not increased 2-13 years after BEP chemotherapy (Martin, 1998). Bahadur et al. (2005) reported that waiting 1.5-2 years may allow for a sufficient turn over of cells to expel any mutagenic effect but then recommended waiting a more practical 6-12 months prior to attempting conception.

There is limited information on the pregnancy outcomes of fathers treated specifically for testicular cancer. The majority of research has focused on issues of fertility and sperm quality.

Hartmann et al. (1999) distributed a questionnaire on fertility and sexual function to patients treated for testicular cancer. Twenty-one out of the 40 couples (53%) that wanted children were able to achieve pregnancy at a median time of 54 months (3-108 months) after treatment ended. Although the authors reported no major birth defects, one child reportedly had cryptorchidism and one had hip dysplasia. Normal development in all children was reported up to a median age of 62 months (1-180 months). In another study, 15 patients treated for testicular cancer by polychemotherapy fathered 20 children with no congenital malformations (van der Kolk, et al., 1990). The mean time from the end of treatment to conception was 39 months (9-82 months). Development of the children was reportedly normal in the children age 3 months-6 years old.

Hodgkin’s Disease
Akin to testicular cancer, several researchers have noted that 30-65% of men have oligospermia or other alterations of semen quality pre-treament with Hodgkin’s disease (Gandini et al., 2003).

Sperm from men treated for Hodgkin’s disease with Novantrone, Oncovin, Vinblastine and Prednisone (NOVP) chemotherapy was examined for frequency of chromosomal abnormalities. Sperm was collected before treatment, shortly after treatment, and 1-2 years after treatment. Researchers examined the chromosomes that are involved in the most common aneuploid syndromes, chromosomes 18, 21, X, and Y. They found a 2-14 fold elevation in abnormal chromosome number in the sample collected shortly after treatment but did not find an increase 1-2 years post treatment (Frias et al., 2003). A similar study found that sperm aneuploid levels were transient and returned to pre-treatment levels  approximately 100 days after the completion of NOVP chemotherapy (Robbins et al., 1997).

Swerdlow et al. (1996) found no excess of stillbirths, low birthweight, chromosome abnormalities, or congenital malformations, and no cancers in 49 offspring of women and men previously treated for Hodgkin’s disease. Sixteen of the children were conceived after chemotherapy, 25 after radiotherapy, and 8 after a combined modality therapy. Sixteen of the children had been conceived less than 5 years after treatment. Additionally, there were no major or minor birth defects among 26 offspring of male patients treated for Hodgkin’s disease (Aisner et al., 1993). The median treatment free interval at the birth of the child was 8 years (1.25-16).

The previous studies with testicular cancer and Hodgkin’s disease illustrate that men should not be counseled that infertility is a definite result of treatment. Cryopreservation of semen prior to treatment is also recommended. Depending on the post thaw semen quality, patients can be advised whether additional fertility assistance procedures should be utilized. A follow-up study on 29 patients did not find a difference in successful pregnancy based on the malignancy type which led to the sperm cryopreservation (Agarwal et al., 2004). If cryopreservation is not performed and normal
fertility does not resume, in-vitro fertilization with single sperm (ICSI) could be attempted post treatment.

Methotrexate
Methotrexate interferes with DNA synthesis and cellular replication. In this way it is effective against malignant cells and cellular proliferation in cancerous tissues. Methotrexate is becoming increasingly prescribed for rheumatoid arthritis and psoriasis. The exposure of fetuses to methotrexate when ingesting directly by women increases the chance for birth defects when taken during the first trimester (and particularly between 8-10 weeks gestation) when the limbs and skull are still forming.

Oligospermia has been seen in individuals taking methotrexate as part of cancer treatment. Whether low dose monotherapy methotrexate taken for rheumatoid arthritis or psoriasis reduces male fertility is still not clear. Reversible sexual dysfunction was reported in three men with rheumatoid arthritis who were treated with weekly doses of 12.5 mg methotrexate. However, two case series published in the 1970s found no change in sperm concentration, motility or quality in 22 men treated with methotrexate for psoriasis (French and Koren, 2003).

While specific studies have not been performed, there are no reports of genetic disease following paternal methotrexate exposure. Therefore, the concern remains theoretical. 6-Mercaptopurine (6-MP) Azathioprine (Imuran) is metabolized to 6-mercaptopurine (6-MP). Both 6-MP and azathioprine appear to work by inhibition of nucleic acid synthesis. 6-MP has been used as a chemotherapeutic agent and in some transplant recipients. These agents are now commonly prescribed to treat inflammatory bowel disease (IBD). Long-term 6-MP treatment in male mice did not impair sperm production or sperm morphology or increase congenital anomalies in offspring (Ligumsky et al., 2005). However, a significantly high rate of embryonic resorption was observed, which the authors postulated could indicate occult sperm damage.

Semen was collected from 23 patients with IBD that had been on azathioprine treatment for at least 3 months. This small study showed no negative association between azathioprine therapy and semen quality (Dejaco et al., 2001). Human paternal studies have yielded conflicting findings in regards to adverse pregnancy outcomes. Studies that have suggested an increased risk for congenital anomalies have been criticized for their small sample size, methodology, and timing contradictions.

Congenital anomalies were seen in 3.3% of 273 pregnancies and 4.8% of 42 pregnancies in male renal or cardiac transplant recipients taking azathioprine (Polifka et al., 2004). These rates are similar to the baseline risk of congenital malformation in the general population. 6-MP use was analyzed in relation to birth outcomes in fathers with IBD. Thirteen pregnancies that had been conceived within 3 months of 6-MP use were compared to a control group consisted of 37 pregnancies that had been conceived at least 3 months after 6-MP use. There were two birth defects and two miscarriages in the infants whose fathers were using 6-MP at the time of conception, and no birth defects and one miscarriage in the group with more remote use.

Based on the better outcomes with the remote use, the authors suggested that males should stop this medicine and wait three months before attempting to conceive. (Rajapakse et al., 2000). However, the rate of birth defects and miscarriage in the control group were lower than expected. Due to the small sample, the results may reflect chance.

A second study in 2003 found no increased risk of birth defects. The authors identified 37 pregnancies in which men were exposed to 25-175 mg 6-MP at the time of conception, 44 pregnancies in which men stopped taking 6-MP at various times prior to conception, and 73 pregnancies in which there was no exposure to 6-MP (Francella et al., 2003). There was one birth defect in the group with exposure at conception, three birth defects in the group with exposure prior to conception (10 months earlier, 3 years earlier, and 4 years earlier), and two birth defects in the group with no exposure. These authors concluded that there is no need for males to discontinue the medicine.

A third study suggested an increased risk for congenital anomalies. Norgard et al. (2004) used a Denmark population database to identify 54 children fathered by men who had filled a prescription for either 6-MP or azathioprine at any point before conception. Four of the 54 children (7.4%) had congenital abnormalities, compared with 2,334 of the 57,195 children (4.1%) fathered by men who had not filled such a prescription. The congenital abnormalities in the exposed group included polysyndactylia; esophageal atresia; hydronephrosis with megaloureter; and a ventricular septal defect. There was no known underlying single gene or chromosomal abnormality.

Additionally, the odds ratio was not statistically significant between the exposed and control group outcomes. Also, the timing of the last prescription for the drugs and conception among the 4 cases with abnormalities ranged from 9 months to 38 months, also casting doubt on the causative effect of the medications upon subsequent birth defects (Cohen, 2004). In fact, none of the 19 pregnancies for which prescriptions for either drug were filled within 3 months of conception resulted in congenital abnormalities.

The question of an increased risk of congenital abnormalities in pregnancies fathered by men on azathioprine or 6-MP based on the available studies should be viewed with skepticism and the benefit to male health kept in mind (Cohen, 2004). Additional studies on larger populations are still needed.

Summary
All men requiring chemotherapy should be offered cryopreservation of sperm before cancer treatment. Individuals receiving cancer treatment that can affect male fertility and sperm quality should be counseled about appropriate contraception for the duration of treatment since infertility cannot not be assumed. It is advisable that couples wait at least 3 months (the time period of one complete spermatogenesis cycle) to one year before attempting conception. Banked sperm may reduce the risk of genetic abnormalities that theoretically could be induced in the stem germ/spermatogonia cells and thus never be eliminated from the mature sperm cells. However, couples should be counseled that there is no evidence of an increased risk for genetic disease in the offspring and that a waiting period is based only on theoretical concerns.

References
· Agarwal A, et al. (2004) Fertility after cancer: a prospective review of assisted reproductive outcome with banked semen
specimens. Fertility and Sterility 81(2):342-348.
· Aisner J, et al. (1993) Pregnancy outcome in patients treated for Hodgkin’s disease. J Clin Oncol 11:507-512.
· Bahadur G, et al. (2005) Semen quality before and after gonadotoxic treatment. Human Reproduction 20(3):774-771.
· Boice JD, et al. (2003) Genetic effects of radiotherapy for childhood cancer. Health Phys 85(1):65-80.
· Brent R (2005) The mutagenic and oncogenic risks of preconception drug, chemical and radiation exposure to male and female gonadocytes. OTIS 18th International Conference.
· Byrne J, et al. (1998) Genetic disease in offspring of long-term survivors of childhood and adolescent cancer. Am J Hum Genet 62:45-52.
· Cohen RD (2004) Sperm, sex, and 6-MP:The perception on conception. Gastroenterology 127(4):1263-1264.
· Davis DL, et al. (1992) Male-mediated teratogenesis and other reproductive effects: biologic and epidemiologic findings and a plea for clinical research. Reproductive Toxicology 6:289-292.
· Dejaco C, et al. (2001). Azathioprine treatment and male fertility in inflammatory bowel disease. Gastroenterology 121(5): 1048-53.
· Francella A, et al. (2003) The safety of 6-mercaptopurine for childbearing patients with inflammatory bowel disese: a
retrospective cohort study. Gastroenterology124:9-17.
· French A and Koren G (2003) Effect of methotrexate on male fertility. Canadian Family Physician 49:577-578.
· Frias S, et al (2003) NOVP chemotherapy for Hodgkin’s disease transiently induces sperm aneuploidies associated with the major clinical aneuploidy syndromes involving chromosomes X, Y, 18, 21. Cancer Res 63(1): 44-51.
· Gandini L, et al. (2003) Testicular cancer and Hodgkin’s disease: evaluation of semen quality. Human Reproduction 18(4):796-801.
· Hartmann JT, et al. (1999) Long-term effects on sexual function and fertility after treatment of testicular cancer. Br J Cancer 80(5-6):801-807.
· Ligumsky M, et al. (2005) Effects of 6-mercaptopurine treatment on sperm production and reproductive performance: a study in male mice. Scand J Gastroenterol 40(4):444-9.
· Martin R (1998) Human sperm chromosome complements in chemotherapy patients and infertile men. Chromosoma 107(6-7): 523-7.
· Martin R, et al (1999) Analysis of sperm chromosome complements before, during, and after chemotherapy. Cancer Genet Cytogenet 108(2): 133-6.
· Norgard B, et al., (2004) The risk of congenital abnormalities in children fathered by men treated with azathioprine or 6-MP before conception. Aliment Pharm Ther 19:679–685.
· Petersen PM, A et al (1994) Dose-dependent impairment of testicular function in patients treated with cisplatin-based
chemotherapy for germ cell cancer. Ann Oncol 5(4): 355-8.
· Polifika JE and Friedman JM (2002) Teratogen update: Azathioprine and 6-mercaptopurine. Teratology 65:240-261.
· Rajapakse RO, et al. (2000) Outcome of pregnancies when fathers are treated with 6-mercaptopurine for inflammatory bowel disease. Am J Gastroenterol 95(3): 684-8.
· Robbins WA, et al. (1997) Chemotherapy induces transient sex chromosomal and autosomal aneuploidy in human sperm. Nat Genet 16(1): 74-78.
· Swerdlow AJ, et al. (1996) Fertility, reproductive outcomes, and health of offspring, of patients treated for Hodgkin’s disease: an investigation including chromosome examinations. Br J Cancer 74(2):291-296.
· Van der Kolk BM, et al. (1990) Children born after their fathers had been treated with chemotherapy for testicular cancer. Eur J Obstet Gynecol Reprod Biol 34(1-2):167-170.
· Winther JF, et al. (2004) Chromosomal abnormalities among offspring of childhood cancer survivors in Denmark: a
population based study. Am J Hum Genet 74: 1282-5.
· Wyrobek AJ, et al. (2005) Relative susceptibilities of male germ cells to genetic defects induced by cancer chemotherapies. J Natl Cancer Inst Monogr 34:31-35.

Contributors
Olivia Hess, BS
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

Methylmercury and Pregnancy

Volume 12, No. 3, July 2005

Mercury is a naturally occurring substance that exists in a variety of chemical forms.

Elemental (inorganic) mercury is found in thermometers and dental amalgams. Mercury vapor (inorganic) is produced by the burning of fossil fuel, or by volcanic eruption, and is dispersed widely in the atmosphere. It is returned to the Earth’s surface via rainwater and undergoes biosynthesis in aquatic environments to become organic mercury, also called methyl mercury (MeHg). Small fish consume the deposited MeHg and bioaccumulation occurs up through the food chain. This issue of Risk Newsletter focuses on the reproductive risks of MeHg mercury. See Risk Newsletter Volume 8 (3) for a reproductive review regarding elemental mercury.

Historical Perspective
High doses of MeHg can cause harmful neurological effects. In adults, MeHg acts on the central nervous system causing an initial sensation of “pins and needles,” which may further progress to ataxia, dysarthria, constriction of the visual fields, and hearing loss. The phenotype and insight into effects on fetal development have been documented following various adult MeHg poisonings.

In 1953 it was reported that villagers in Minimata, Japan developed neurological disease following high level exposure to MeHg. A factory located near Miniamata Bay had dumped their inorganic mercury waste into the bay. It was eventually estimated that the amount of mercury dumped exceeded 100 tons (Tedeschi, 1982). Fish consumed the MeHg and the villagers’ diet largely consisted of the contaminated fish. Residents experienced ataxia, sensory/motor impairment, speech impairments and visual disturbances (Tedeschi, 1982). There were about 380 confirmed cases of MeHg poisoning with 52 fatalities (Tedeschi, 1982). Additionally, pregnant women consuming the contaminated fish gave birth to infants with cerebral palsy, mental retardation, and seizures. This condition is sometimes referred to as Congenital Minimata Disease.

In 1971, another epidemic of MeHg poisoning occurred in Iraq when grain treated with MeHg as an antifungal agent was mistakenly ground into bread and eaten directly instead of being planted (Moienafshari, 1999). Some 6000 persons were admitted to the hospital and there were over 450 deaths (Clarkson, 2002, Tedeschi, 1982). Again the offspring of pregnant women were noted to have higher rates of microcephaly, cerebral palsy, and sensory impairments.

Based on this Iraqi population, a 5% risk for neurological damage was suggested following mercury levels of 10-20 ug/g in maternal hair. This should be viewed as a tentative estimate since many children born to women with hair levels >100 ug/g had normal outcomes and a more sophisticated re-analysis suggested no adverse neurological damage with maternal hair < 80 ug/g (Bellinger, 2003). In comparison, the average United States diet results in maternal hair concentrations of less than 1 ug/g (McDowell, 2004, Davidson, 2005)

In both the Japan and Iraq populations, neurological disease was noted in children born to mothers with only mild, transient paresthesias (or no signs at all) suggesting a heightened fetal sensitivity (Bellinger, 2003).

Recent population studies
Populations in archipelago environments in which diets consist largely of fish have made possible longitudinal prospective studies of MeHg exposure in non-poisoning situations. While levels of MeHg were comparable between studies, disparate results have made it difficult to draw conclusions regarding safety threshold levels.

The Faroe Islands cohort included 917 children enrolled between 1986 and 1987 who were then evaluated at 7 years of age for cognitive deficits (Grandjean et al., 1997). Pilot whale meat is a staple food in this population and the main source of MeHg exposure. Mercury concentrations were measured in both cord blood and maternal hair (mean 4.27 ug/g). All analyses were carried out using MeHg concentrations from cord blood.

Clinical examination and neurophysiolocial testing did not reveal any obvious mercury related anomalies. However, following multiple regression analysis 9 out of 20 neuropsychological measures showed mercury-associated deficits. Deficits were most apparent in measures of language, attention, and memory, with fewer effects seen in domains of visuospatial and motor functioning.

Of note, they saw effects even when they excluded children with maternal hair mercury concentrations above 10 ug/g. Fish in the Seychelle Islands have similar MeHg concentrations as do fish in the United States, however, fish consumption is greater with an average of 12 weekly fish meals versus one or less for the United States (Myers et al., 2003). The Seychelles archipelago cohort consisted of 643 mother-child pairs enrolled between 1989 and 1990. Prenatal exposure to MeHg was detemined by measuring total Hg in maternal hair during pregnancy. Children were followed for 9 years and evaluated for neurocognitive, language, memory, motor, perceptual-motor, and behavorial functions through specific testing. The authors controlled for caregiver IQ, socioeconomic status, and home environment stimulation. The mean prenatal total MeHg exposure was 6.9 ug/g.

Only 2 of 21 end points were associated with prenatal MeHg exposure, namely, improved scores on the hyperactivity domain of the Connor’s teacher rating scale and decreased performance on the grooved pegboard non-dominant hand in males only. Based on further analysis, these findings were thought to be due to chance. The authors concluded that their study did not support the hypothesis that there is a neurodevelopmental risk from prenatal MeHg exposure resulting solely from ocean fish consumption (Myers et al., 2003).

While both studies are high quality with regard to methodology , there are several possible explanations for the incongruent results between the Faroe and Seychelle Island studies (Davidson, 2005). For example, sources of MeHg exposure differed; whale meat in the Faroe Islands and ocean fish in the Seycelle Islands. Pilot whale meat has much higher levels of mercury than other seafood and the blubber also contains more polycholorinated biphenyls (PCBs) and other chemicals.

Measurement procedures also differed between studies. Cord blood concentrations used in the Faroe Island cohort are indicative of concentrations only in the third trimester, whereas MeHg concentrations in maternal hair used in the Seycelle study indicate exposure throughout pregnancy.

This difference may reflect effects of episodic versus long-term exposure. The disparate results could also lie in population differences in genetic, nutritional, and social-environmental areas (Bellinger, 2003). Daniels et al. (2004) evaluated maternal fish consumption and very early offspring cognitive development in a cohort of 7421 British children born in 1991-1992. Mercury levels were measured in cord tissue in 1054 children. Maternal and child fish intake was assessed by patient questionnaire and developmental tests were performed at 15 and 18 months of age. Maternal fish intake was categorized into rarely or never, once per 2 weeks, 1-3 times per week, and 4 or more times fore week. The authors assumed the each fish meal averaged 4.5 ounces.

Maternal fish intake was associated with increased umbilical cord mercury concentrations but the overall cord mercury levels were low (median 0.01ug/g wet weight) and not associated with developmental outcomes. Maternal fish intake during pregnancy was actually associated with a subtle but consistently higher developmental scores for language comprehension and social activity compared to women who did not eat fish. The association was strongest for women eating fish, 1-3 times per week. The authors concluded that fish intake could subtly enhance early child development. However, it cannot be known from this study whether it was the fish consumption itself or an associated factor, such as an overall better diet or better caregiving, that contributed to the outcome.

FDA Advisory
The March 2004 advisory from the U.S. Food and Drug Administration (FDA) and the Environmental Protection Agency made the following recommendations for women who may become pregnant, pregnant women, nursing mothers, and young children:

·Avoid shark, swordfish, king mackeral, and tilefish (larger fish with longer life spans accumulate the highest levels of MeHg)

·Eat up to 12 ounces (2 average meals) of other cooked fish weekly

·Commonly eaten fish that are low in mercury include but are not limited to canned light tuna, shrimp, salmon, catfish, and pollock

·http://www.cfsan.fda.gov/~frf/sea-mehg.html contains mercury level listing of other fish

·Since albacore (”white”) tuna has more mercury than canned light tuna, when choosing two meals of fish and shellfish, eat up to 6 ounces (one average meal) of albacore tuna per week.

·Check local advisories about the safety of fish caught by family and friends in your local lakes, rivers, and coastal areas. If no advice is available, eat up to 6 ounces (one average meal) per week of fish you catch from local waters, but don’t consume any other fish during that week.

One weeks’ consumption does not significantly alter the body mercury levels so these are “on average” guidelines

Summary
Fish is an important source of protein and omega 3 fatty acids before, during, and after pregnancy. In order to accrue the benefits of fish but maintain lower mercury levels, an average of 12 ounces of certain fish per week is recommended. This FDA recommendation is thought to be conservative given the lack of consensus findings of adverse neurological effects with non poisoning situations. While MeHg can be measured in hair (chronic exposure) or blood (recent exposure), a risk assessment based on the results is unclear so routine pregnancy testing is not recommended.

References
·Bellinger DC (2003) Mercury in fish: implications for pregnant women. Human Teratogens Course: Harvard Medical School .
·Clarkson TW (2002) The three modern faces of mercury. Environmental Health Perspectives 110(1):11-23.
·Daniels JL, et al. (2004) Fish intake during pregnancy and early cognitive development of offspring. Epidemiology 15(4):394-402.
·Davidson PW, Myers GJ, Weiss B (2004) Mercury exposure and child development outcomes. Pediatrics
113(4):1023-1028.
·Grandjean P, et al. (1997) Cognitive deficit in 7-year-old children with prenatal exposure to methylmercury Neurotoxicology and Teratology 19(6):417-428.
·Mahaffey KR (1999) Methylmercury: a new look at the risks. Public Health Reports. 114: 397-415.
·Marsh DO, et al. (1987) Fetal methylmercury concentration in single strands of maternal hair and child effects. Arch
Neurol 44:1017-1023.
·McDowell MA, et al. (2004) Hair mercury levels in the U.S. children and women of childbearing age:reference range data
from NHANES 1999-2000. Environ Health Perspect 112:1165-1171.
·Moienafshari R, Bar-Oz B, and Koren G (1999) Occupational exposure to mercury: what is a safe level? Canadian Family Physician 45:43-45.
·Rustam H, Hamdi T (1974) MeHg poisoning in Iraq a neurological study. Brain 97:499-510.
· Tedeschi LG (1982) The Minamata Disease The American Journal of Forensic Medicine and Pathology 3(4): 335-338.
US Department of Health and Human Services and the Environmental Protection Agency (2004) What You Need to
Know About Mercury in Fish and Shellfish: EPA and FDA Advice For:Women Who Might Become Pregnant
Women Who are Pregnant Nursing MothersYoung Chil dren http://www.cfsan.fda.gov/~dms/admehg3.html

Contributors
Hazel Perry, BS
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

Update on Proton Pump Inhibitors During Pregnancy

Volume 12, No.2, May 2005

GERD
Gastro-esophageal reflux disease (GERD) is a chronic disease characterized by repeated or prolonged exposure of the lining of the esophagus to the acidic contents of the stomach. Common symptoms include heart burn and acid regurgitation. Approximately 22% of pregnant women will suffer from reflux during the first trimester, 39% develop it by the second trimester, and up to 72% of women develop it by the third trimester (Tytgat et al, 2003).

GERD rarely causes serious complications during pregnancy, but symptoms may be unpleasant and require treatment. There is a common treatment protocol for pregnant women outlined in the literature. GERD is typically treated with dietary and lifestyle changes, along with periodic use of over-the-counter calicium and magnesium based antacids (Tytgat el al., 2003). If this approach is not effective, the H2-blocker rantidine is often tried concurrently with antacids (Katz et al., 1998; Richter, 2003). Proton pump inhibitors are generally reserved for more severe cases.

Proton pump inhibitors are effective treatments for conditions such as GERD because they block enzymes in the wall of the stomach that produce acid. A variety of GERD medications were reviewed in the June 1999 (RISK//NEWSLETTER 7(5)); this newsletter serves as an adjunct to that issue.

This newsletter will focur on the proton pump inhibitors omeprazole (Prilosec), lansoprazole (Prevacid), rabeprazole (Aciphex), pantoprazole (Protonix), and esomeprazole magnesium (Nexium).

Omeprazole (brand name: Prilosec)
Omeprazole was approved by the FDA in 1989. Currently, it is the best studied proton pump inhibitor for use during pregnancy. A reproductive study in rats and rabbits found no teratogenic effect in fetal development after administering doses up to 250-500 times greater than the recommended human dose (Lalkin et al., 1998). There was a slight increase in fetal loss at the top doses.

Although there are individual case reports of congenital anomalies, controlled studies have not found an increased risk of major malformations. A prospective cohort study by Lalkin et al. (1998) described 113 pregnancies exposed to omeprazole compared to a control group taking H2- blockers and a control group with no prescription GERD medications. Eighty-five percent of the women took omeprazole during the first trimester, while 15% continued use through delivery.

There were no significant differences in the number of congenital anomalies, miscarriages, mean birthweight, or premature delivery between the three groups. Based on the small sample size, this study had an 80% power to identify a 5-fold increased risk.

Two databases, one from England and the other from Italy, were combined in a study published in 1999 that was designed to assess the incidence of congenital malformations in women who had received a prescription for an acid-suppressing drug including omeprazole during the 1st trimester (Ruigomez et al., 1999). Omeprazole was taken in 134 pregnancies, resulting in 139 live births. There was no significant increase in the rate of malformations.

A review of the Swedish medical birth registry by Kallen (2001) identified 955 infants exposed to omeprazole during pregnancy. Eighty-six percent of the infants were exposed only during the first trimester. Specific data on dose and timing was not available. The authors found no significant increase in congenital anomalies, low birth weight, low Apgar scores, or perinatal survival.

Most recently, a multicenter prospective case controlled study by Diav-Citrin et al. (2005) identified 295 women exposed to omeprazole during pregnancy. Sevnty-nine percent of exposures occurred during the first trimester only. The median dose was 20 mg (20-40 mg) and median duration of use was 22 days (4-47 days). The incidence of congenital anomalies in the study population was not greater than the observed frequency in the control group. This sample size was noted to have an 80% power to detect a 2.72-fold increase of major malformations. There were also no significant differences in the rates of miscarriage, ectopic pregnancies, stillbirths, or preterm deliveries. There was a significant 60 g reduction in the median birth weight in the exposed group which has not been replicated in other studies.

Brunner et al. (1998) identified nine cases of maternal exposure to omeprazole: four women were treated during the first trimester and of those, three women continued taking omeprazole until delivery. Five women were treated during the third trimester only. There were no congenital anomalies reported. Brunner et al. followed-up with the children on an average of 5 years (2-12 years) and reported normal development in all nine children.

Lansoprazole (brand name: Prevacid)
Lansoprazole was approved by the FDA in 1995. A reproductive study in rats and rabbits by Schardein et al. (1990) found no adverse effect on fertility and no increased incidence of congenital anomalies after administering doses 16 to 40 times the recommended human doses.

The largest study sample size on lansoprazole consisted of only 62 pregnancies exposed to lansoprazole, 55 of which were during the first trimester (Diav-Citrin et al., 2005). The median dose was 30 mg (30-60 mg) and the median duration of use was 14 days (7-32 days). The incidence of congenital anomalies in the study population was not greater than the observed frequency in the control group. This sample size had an 80% power to identify a 4.75-fold increase risk of major malformations. There were no significant differences in the rates of miscarriages, ectopic pregnancies, stillbirths, or the rate of preterm deliveries.

Rabeprazole (brand name: Aciphex)
Rabeprazole was approved by the FDA in 1999. In preclinical studies reported by the manufacturer, no teratogenic effects were seen in rats or rabbits at 8-13 times the human dose (Product information, Aciphex, 1999). Administered doses of 195 times the human dose during late pregnancy and lactation in rats decreased weight gain of the pups. Currently there are no studies evaluating the safety of rabeprazole use during pregnancy in humans.

Pantoprazole (brand name: Protonix)
Pantoprazole was approved by the FDA in 2000. A reproductive study by the manufacturer in rats and rabbits found no adverse effects in fetal development after administering doses 16 to 88 times greater than the recommended human dose. (Product information, Protonix, 2001).

Diav-Citrin et al. (2005) prospectively followed 53 pregnancies exposed to pantoprazole. Forty-seven exposures occurred during the first trimester. All women took 40 mg omeprazole daily for a median duration of 14 days (7-23 days). There were no significant differences in the rates of miscarriages, ectopic pregnancies, stillbirths, or rate of preterm deliveries. The incidence of congenital anomalies in the study population of 48 infants was not greater than the observed frequency in the control group. This study had an 80% power to identify a 4.9-fold increase risk of major malformations.

Esomeprazole (brand name: Nexium)
Esomeprazole was approved by the FDA in 2001. An animal study by the manufacturer found no adverse effects in fertility or embryo development after administering oral doses to rats up to 57 times the human dosage and oral doses to rabbits up to 35 times the human dose (Product information, Nexium, 2001).

There are no human studies specific to esomeprazole. However, due to the chemical similarities to omeprazole, studies on the latter should have some relevance. Omeprazole is a racemate, meaning it contains two compounds (isomers) with the same chemical components but different spatial dispositions which leads to different pharmacological properties (Kendall 2003). Esomeprazole is one of isomers of omeprazole.

Summary
During pregnancy, dietary changes and limited use of antacids and/or rantidine is typically the primary treatment for GERD. Proton pump inhibitors are typically reserved for pregnant women with moderate/severe gastrointestinal symptoms for which the former treatments are ineffective. As a class, current data does not suggest that proton pump inhibitors represent an increased risk for major malformations. However, only omeprazole has several human studies for better confirmation. Chronic use throughout pregnancy is not well studied.

References
· Briggs, G et al. (2002) Drugs in Pregnancy and Lactation, 6th Edition. MD: Lippincott Williams & Wilkins.
· Brunner G, et al. (1998) Omeprazole for peptic ulcer disease in pregnancy. Digestion 59: 651-654.
· Diav-Citrin O, et al. (2005) The safety of proton pump inhibitors in pregnancy: a multicenter prospective controlled study. Aliment Pharmacol Ther 21(3): 269-275.
· Kallen BA. (2001) Use of omeprazole during pregnancy: no hazard demonstrated in 955 infants exposed during pregnancy. Eur J Obstet Gynecol Reprod Biol 96(1): 63-68
· Kendall MJ (2003) Review article: esomeprazole-the first proton pump inhibitor to be developed as an isomer. Aliment Pharmacol Ther 17 (Suppl.1)1-4.
· Katz PO, Castell DO. (1998) Gastroesophageal reflux disease during pregnancy. Gastroenterol Clin North Am 27(1): 153- 167.
· Lalkin A, et al. (1998) The safety of omeprazole during pregnancy: a multicenter prospective controlled study. Am J Obstet Gynecol 179:727–30.
· Nikfar S, et al. (2002) Use of proton pump inhibitors in pregnancy ad rates of major malformations: a metaanalysis. Dig Dis Sci 47(7): 1526-1529.
· Product information (1999) Aciphex. Eisai.
· Product information (2001) Nexium. Astra Zeneca.
· Product information (2001) Protonix. Wyeth-Ayerst.
· Rayburn W, et al. (1999) Antacids vs. antacids plus non-prescription ranitidine for heartburn during pregnancy. Int J Gynecology & Obstetrics 66: 35-37.
· REPROTOX online teratology database www.reprotox.org
· Richter JE. (2003) Gastroesophageal reflux disease during pregnancy. Gastroenterol Clin North Am 32(1): 235-26.
· Ruigomez A et al. (1999). Use of cimetidine, omeprazole, and ranitidine in pregnant women and pregnancy outcomes. Am J Epidemiol 1999;150:476–81.
· Schardein JL, et al. (1990) Reproductive and developmental toxicity studies of lansoprazole (AG-1749) in rats and rabbits. Yakuri Chiryo 18 (Suppl 10): 119-129.
· Tytgat GN, et al. (2003) Contemporary understanding and management of reflux and constipation in the general population and pregnancy: a consensus meeting. Aliment Pharmacol Ther 18: 291-301.
· Wilton LV, et al. (1998) The outcomes of pregnancy in women exposed to newly marketed drugs in general practice in England. Br J Obstet Gynaecol 105: 882-889.

Contributors
Dania Stachiw, BS
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

The Newer SSRI Anti-Depressants and Pregnancy

Volume 12, No. 1, April 2005

The prevalence of clinical depression during pregnancy has been estimated to be 7-12% by a recent meta-analysis. (Bennett et al., 2004). Additionally, in over 3000 obstetric patients screened, 20% had high scores on Centre for Epidemiological Studies Depression Scale (Bonari et al., 2004).

Therefore, it is not surprising that the Illinois Teratogen Information Service receives many requests for information regarding pharmacological treatment of depression during pregnancy.

The decision of whether or not to keep a woman on medication during pregnancy is complex. The many possible adverse effects of untreated depression must be considered. There is a growing body of literature on an association of obstetrical complications with untreated maternal depression. Women with untreated depression are also more likely to use alcohol, tobacco, and illicit drugs (Zuckerman et al., 1989). They may be less able to motivate themselves to attend prenatal appointments or follow medical advice and have poor nutrition (Bonari et al., 2004). The risk of self injury and suicide is a real concern. Thus, it is essential to consider each case individually prior to making medicine changes.

While past newsletters (March 1995 and 1999) have discussed the selective serotonin reuptake inhibitors (SSRIs), Prozac, Zoloft, and Paxil, this newsletter reviews current literature regarding prenatal exposure to the newer SSRIs, Lexapro and Celexa. NonSSRIs antidepressants,

Wellbutrin and Effexor, are also discussed.

Lexapro (escitalopram)
Escitalopram is an SSRI used in the treatment of depression. Escitalopram, marketed as Lexapro, is the active isomer of citalopram, thus the two compounds are chemically similar.

Congenital anomalies
Unpublished animal studies examining the effects of prenatal exposure to escitalopram have been conducted by the manufacturer (Forrest Labs). No increase in congenital anomalies was seen at doses 75X the maximum recommended human dose (MRHD) in rats. Decreased fetal body weights, delays in ossification, slightly increased offspring mortality, and signs of maternal toxicity were observed at the higher doses.

To date there have been no human studies examining the effects of prenatal exposure to escitalopram. There is human data, however, about citalopram use in pregnancy, the results of which should have some application to escitalopram.

Celexa (citalopram)
Citalopram, marketed as Celexa, is an SSRI used to treat depression. Due to the relatively recent marketing of this medication (approved by the FDA in 1998), data examining the effects of prenatal exposure, particularly long term development, is limited.

Congenital anomalies
Unpublished experimental animal studies examining the effects of prenatal exposure to citalopram have been conducted by the manufacturer (Forrest Labs). Teratogenic effects (cardiovascular and skeletal defects) were only observed at the highest doses when maternal toxicity symptoms were present. Lowered birth weights and increased offspring mortality were also present at the higher doses. No adverse effects were seen with rabbit studies with doses 5X MRHD.

Two small prospective human studies have been performed examining birth outcomes following prenatal exposure to citalopram. One study prospectively followed a group of 10 women taking citalopram 20-40mg/day throughout pregnancy and one woman taking citalopram starting in the second trimester (Heikkinen et al., 2002). These women were being treated for either depression or panic disorders. There were no major malformations present and no differences in the Apgar scores or birthweight compared to a control group.

Additionally, a larger prospective study from a drug recording program of the Swedish Medical Birth Registry reported the pregnancy outcomes of 531 infants prenatally exposed to SSRIs, 365 of which were exposed to citalopram (Ericson et al., 1999). This registry identifies early pregnancy exposures, but timing and dosing were poorly specified. Of those infants exposed to citalopram, there was no increase in major malformations and no pattern to any of the birth defects.

The authors also noted that due to two reports to the FDA, they specifically checked but did not find any cases of optic nerve hypoplasia, although the infants were evaluated only in the perinatal period. They did find a small but significant increase in prematurity (OR 1.6) in women taking any SSRIs. It is not clear what role the medicine, maternal condition, or life style factors associated with the maternal condition played.

Neonatal withdrawal
Nordeng et al. (2001) described 5 possible cases of neonatal withdrawal with various SSRIs, including one case of an infant whose mother took 20mg/day of citalopram from months 5-7 and an increased dose of 30mg/day from months 7-9. The infant was born at term and had light abstinence symptoms with increased tonus in his extremities and neck, and was jittery. All symptoms, with the exception of tonus, resolved within seven days of birth (Nordeng et al., 2001). No medical treatment was needed.

Laine et al. (2003) prospectively followed 20 infants exposed to the SSRIs citalopram (N=10) and Prozac (N=10) compared to control group of healthy women not receiving psychotropic medication. While they found a 4-fold increase in serotonergic symptoms (tremor, restlessness and rigidity) in SSRI-exposed infants ages 1-4 days, there were no significant differences in symptom scores when only the citalopram group was compared to the controls. No differences were found in vital signs such as blood pressure, heart rate and body temperature between the two groups, with the exception of a statistically significant increase in heart rate in the SSRI group at 2 weeks of age. No specific medical treatment was needed.

Long term development
Psychotropic medications alter neurotransmitter levels in the maternal brain, thus there is a theoretical risk that they can also alter the developing fetal brain. These brain alterations could potentially lead to behavioral or learning deficiencies.

In the study by Heikkinen et al. 2002 (discussed previously), 11 infants who were exposed to citalopram in utero were followed until age 1 year. The body weights of all infants were normal at one year, as was the neurological development. One child could not walk at the age of 1 year, but the neurological status of this child was evaluated as normal 6 months later (Heikkinen et al., 2002). Due to the limitation of the small sample size and limited follow up time, further investigation in this area is necessary.

Wellbutrin/Zyban (bupropion) Bupropion is an aminoketone used both as an antidepressant (Wellbutrin) and as an aid in smoking cessation (Zyban).

Congenital anomalies
One study in rabbits only saw an increase in skeletal anomalies and delayed ossification at the highest dose when maternal toxicity was present (Tucker, 1983). This same study reported that high doses given to pregnant rats produced maternal toxicity, but that no congenital anomalies were found in the offspring.

The majority of human data on pregnancy exposure is available through a bupropion pregnancy registry maintained by the manufacturer, GlaxoSmithKline. The registry prospectively collected pregnancy exposure information and outcomes since September 1997. Controls were not used. The current update of this registry (February 2004) indicated that 534 pregnancy outcomes have been prospectively analyzed with 354 live births following first trimester exposure. Of these first trimester exposures, there were 12 pregnancy outcomes resulting in birth defects (3.4%), which is not higher than the general population.

However, of the 12 birth defects, 7 were isolated heart defects. Therefore there is continued study to assess an association specifically with cardiac defects.

Additionally, a prospective controlled study found no increase in the rate of malformations in 136 women taking bupropion in the first trimester (Chun-Fai-Chan et al., 2005). Forty-five women took bupropion throughout pregnancy. There were 72 livebirths and no major malformations reported. Due to the small sample size this study had a 80% power to identify a 5-fold increase risk for malformations.

There were also no differences in birth weight, gestational age at birth, or stillbirth compared to the controls. There was a significant increase risk for miscarriage among women taking any antidepressant or taking bupropion (12.3-15.4%) compared to a control group without depression (6.7%). In addition to the medicines, this latter finding could reflect factors with the maternal depressive condition or simply reflect an unusually low miscarriage rate in control women.

Neonatal withdrawal
No reports of neonatal withdrawal have been published.

Long term development
There are at present no published studies examining the long term effects on development in children prenatally exposed to bupropion.

Effexor (venlafaxine)
Venlafaxine is a bicyclic antidepressant marketed as Effexor.

Congenital anomalies

Unpublished animal data available from the manufacturer (Wyeth) reported no increase in malformations in the offspring of rats given up to 11 times and rabbits given up to 12X MHRD. Decreased weight and viability were noted in offspring at 10X MHRD in rats. Another study by da-Silva et al. produced similar results in rats prenatally exposed to venlafaxine (da-Silva et al., 1999).

They found no increased risk for congenital malformations, but did find a slight decrease in birth weight of litters exposed to venlafaxine. Data obtained from the U.K. Drug Safety Research Unit included pregnancy outcomes for 26 live births in women who took venlafaxine during pregnancy (Einarson et al., 2001). No major malformations were reported.

A larger study by Einarson et al. (2001) prospectively collected information about birth outcomes of 150 women exposed to venlafaxine during pregnancy, 126 of which used the medication during the first trimester and 34 throughout the pregnancy.

Seventy percent of the woman took 75mg venlafaxine (range: 37.5-300mg). The authors compared pregnancy outcomes of this group to two control groups, women who used SSRI’s during pregnancy and women who did not take antidepressants. Pregnancy outcomes such as the number of live births, spontaneous abortions, preterm delivery, birth weight and major malformations were examined. The pregnancy outcomes were not statistically different between the three groups for any of these factors. The results from this study are promising, but are limited by the small sample size which provides only an 80% power to detect a 4-fold increase in malformations.

Neonatal withdrawal
The WHO Collaborating Centre for International Drug Monitoring in Sweden described 17 reports of neonatal withdrawal syndrome, of which only 6 were regarded as potentially related (Sanz et al., 2005).

An individual case report in a German journal described an infant with restlessness, hypertonia, jitteriness, irritability and poor feeding (de Moor et al., 2003). The diagnosis of neonatal withdrawal was further suspected when there was a temporary improvement after administration of low dose (1 mg) venlafaxine. After 8 days, with no further treatment, the infant’s symptoms resolved.

Long term development
The long term effects on development in children prenatally exposed to venlafaxine has not been studied.

Summary
Reproductive studies to do not suggest an increase in congenital anomalies for any of the four medications discussed, although the studies are still limited by their small sample sizes. Studies with sample sizes less than 150 women allow at most a 4 fold detection of increased risk (Einarson and Einarson, in press). Further evaluation of a possible association with bupropion and cardiac defects is still needed.

Exposure to antidepressants which inhibit serotonin reuptake during the third trimester of pregnancy carries the risk of a neonatal withdrawal/serotonergic syndrome. Reported symptoms are non specific and typically self limiting. Symptoms most commonly reported include agitation, irritability, hypotonia, hypertonia, hyperreflexia, drowsiness, persistent crying, and sucking problems
(Prescrire International, 2004). Small nonblinded case series suggest such findings occur in 20-30% of infants with third trimester exposure to the older SSRIs (Prozac, Paxil, and Zoloft) compared to 6-9% of control infants (Prescrire International, 2004).

Individual case reports of withdrawal-like symptoms also exist for Celexa and Effexor. However, discontinuing antidepressants near delivery is controversial due to maternal and infant adverse effects with postpartum depression. Longterm studies on whether there are any effects on neurobehavioral outcomes are absent. Currently there only studies on early childhood development for Prozac and the tricyclic antidepressants.

It should be noted that stopping medication is not a “no risk” option since there are concerns to a pregnancy with untreated maternal depression. The severity of maternal symptoms should help dictate medicine use during pregnancy.

References
Bennett, et al. (2004) Prevalence of depression during pregnancy: systematic review. Obstet Gynecol 103(6):698-709.

Bolton HL, et al. (1998) Incidence and demographic correlates of depressive symptoms during pregnancy in an inner London population. J Psychosom Obstet Gynaecol 19(4):202-9.

Bonari L, et al. (2004) Perinatal risks of untreated depression during pregnancy. Can J Psychiatry 49(11)726-735.

Chun-Fai-Chan, et al. (2005) Pregnancy outcome of women exposed to bupropion during pregnancy: A prospective comparative study. Am J Obstet Gynecol 192(3):932-936.

da-Silva et al. (1999) Postnatal development of rats exposed to fluoxetine or venlafaxine during the third week of pregnancy. Braz J Med Biol Res 32(1):93-8.

de Moor RA, et al. (2003) [Withdrawal symptoms in a neonate following exposure to venlafaxine during pregnancy] Ned Tijdschr Geneeskd. 12;147(28):1370-2.

Ericson A, et al. (1999) Delivery outcome after the use of antidepressants in early pregnancy. Eur J Clin Pharmacol 55:503-508

Einaron TR and Einarson A (In press) Newer antidepressants in pregnancy and rates of major malformations: a meta-analysis of prospective comparative studies Pharmacoepidemiology and Drug Safety.

Einarson A, et al. (2001) Pregnancy outcome following gestation exposure to venlafaxine: a multicenter prospective controlled study. Am J Psychiarty 158:1728-1730.

Forest Laboratories (2005) Product information.

GlaxoSmithKline Bupropion Pregnancy Registry (2004) Research Triangle Park, NC.

Heikkinen T, et al. (2002) Citalopram in pregnancy and lactation. Clin Pharmacol Ther 72(2):184-91.

Kessler R, et al. (1993) Sex and depression in the National Comorbidity Survey I: Lifetime prevalence, chronicity and recurrence. Journal of Affective Disorders 29:85-96.

Laine K, et al. (2003) Effects of exposure to selective serotonin reuptake inhibitors during pregnancy on serotonergic symptoms in newborns and cord blood monoamine and prolactin concentrations. Arch Gen Psychiarty 60:720-726.

Nordeng, et al. (2001) Neonatal withdrawal syndrome after in utero exposure to selective serotonin reuptake inhibitors. Acta Paediatr 90:288-291.

Prescrire International (2004) Neonatal complications after intrauterine exposure to SSRI antidepressants. 13(71)103-104.

Sanz EJ, et al. (2005) Selective serotonin reuptake inhibitors in pregnant women and neonatal withdrawal syndrome: a database analysis. Lancet 365:482-487.

Tucker WE (1983) Preclinical toxicology of bupropion: An overview. J Clin Psychiatry 44:60-62.

Zuckerman, et al. (1989) Depressive symptoms during pregnancy: relationship to poor health behaviors. Am J Obstet Gynecol 160:1107-1111.

Contributors
Michelle Martin, BS
Genetic Counseling Intern
Jennifer Sloan, PhD
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

The New Antiepileptic Drugs and Pregnancy

Volume 11, No.3, August 2004

This issue of Risk Newsletter explores the current data on the following second generation antiepileptic drugs (AEDs): Lamictal, Trileptal, Topamax, and Neurontin.

About 1 in every 200 pregnant women has epilepsy, and up to 35% of women with epilepsy have increased seizure activity during pregnancy (AAN, 1998). The latter may be due to changes in sex hormones, metabolism, sleep patterns, or medication compliance (Morrell, 2003). The majority of these pregnancies have normal outcomes; however, there are specific maternal and fetal concerns related to uncontrolled seizures. While studies are not conclusive, seizures during pregnancy have been associated with such obstetrical complications as prematurity and stillbirth (Morrell, 2003, AAN, 1998). Seizures can result in maternal and fetal hypoxia and acidosis, which potentially could lead to neurological damage or be life-threatening. Therefore, treatment during pregnancy is typically instituted when a woman has a history of seizures within the past 24 months. If therapy is to be withdrawn during pregnancy, it is ideally attempted 6 months prior to conception to allow time to evaluate for seizure recurrence (AAN, 1998).

The risk for congenital malformations ranges from 4-10% following monotherapy treatment with older, “first generation” AEDs (Dolk and McElhatton, 2002). This is a 2- to 3-fold increased risk compared to the general population. Although this increased risk was once thought to be due to the underlying maternal seizure disorder, it is now largely attributed to the use of AEDs (Holmes et al., 2001). A woman’s risk for congenital anomalies increases with polytherapy (AAN, 1998). Therefore, monotherapy treatment with the medicine most effective for the specific seizure type is preferred.

Second generation AEDs have been available since 1993. Unfortunately, these medications have not been well studied and, consequently, risks to the fetus are largely unknown. Of studies that have been published, major malformations is the only outcome variable evaluated. Studies have not been conducted to determine if there are any patterns of minor malformations or developmental impairment (i.e., anticonvulsant embryopathy).

Lamictal (generic name: lamotrigine)
For second generation AEDs, the largest dataset on pregnancy outcomes is available for lamotrigine (LTG), a medicine chemically unrelated to the first generation AEDs. While LTG decreases fetal folate levels in rats, it does not decrease human adult blood folate levels (GlaxoSmithKline, 2004). It is not known if human fetal folate levels are altered with LTG or whether higher maternal folic acid supplements (4-5 mg daily) is beneficial for women taking Lamictal.

Marchi et al., (2001) treated pregnant rats during organogenesis with four times the recommended human dose of LTG. Offspring demonstrated low birth weight and altered brain structure, which included increased volume and diameter of the cerebral structure, increased density of the subcortical layer, and ventricle dilation (Marchi et al., 2001). The relevance of these findings to human pregnancy is unknown. Experimental animal studies by the manufacturer did not find an increase in congenital malformations associated with LTG (GlaxoSmithKline, 2004).

Sabers et al., (2004) reported on the outcomes of 147 human pregnancies with various AED exposures. Seventy-four percent of the total group were on monotherapy AED treatment, and 80% of the total group took folic acid supplements, with the majority taking 5 mg daily. Of the total group, 35% were treated with LTG. For monotherapy treatment with LTG, there were no major malformations. There was one case of a ventricular septal defect in a LTG exposed pregnancy also treated with oxcarbazepine.

The largest dataset on LTG exposure during pregnancy has been collected by the manufacturer’s pregnancy registry. This is a voluntary, noncontrolled prospective registry with 20-30% of women lost to follow-up. As of March 2003, the registry had outcomes for 414 first trimester exposures to LTG monotherapy. There were 12 infants with major malformations for a rate of 2.9%. The spectrum of birth defects observed and the proportion of malformations were not different from the baseline risks. This sample size was sufficient to detect, with 80% power, a 1.79-fold increase in the proportion of major birth defects. Additionally, an abstract from the United Kingdom Lamictal registry reported on 390 first trimester pregnancy exposures. There were 8 major malformations for a malformation rate of 2.1%, which was within the rate present in the general population (GlaxoSmith Kline, 2004).

Based on available data, LTG does not appear to significantly increase the chance for congenital anomalies. Long-term data regarding any possible neurobehavorial effect of LTG exposure in utero is not available.

Trileptal (generic name: oxcarbazepine)
There were no significant increases in congenital anomalies when pregnant mice were treated with 20-46 times the human dose of oxcarbazepine (OXC) (Bennett et al., 1996). However, studies conducted by the manufacturer demonstrated increased craniofacial, cardiovascular and skeletal malformations in rats treated during organogenesis with 1.2 times and 4 times the maximum recommended human dose of OXC. In rabbits, there was an increase in fetal loss but no teratogenicity associated with OXC administration at 1.5 times the maximum human dose (Novartis Pharmaceuticals, 2004).

Reports on human pregnancy exposure to OXC are limited and most studies do not specify whether the women were on OXC monotherapy or polytherapy. In one study with 12 pregnancies exposed to OXC during the first trimester, three pregnancies resulted in miscarriage, while nine pregnancies resulted in newborns without structural malformations (Friis et al., 1993).

Another study of 37 women exposed to OXC, alone or in conjunction with other AEDs, demonstrated two cases of ventricular septal defects. One case was exposed to OXC alone and the other was exposed to OXC and LTG, as mentioned previously (Sabers et al., 2004).

Additional reports specifying OXC monotherapy during the first trimester total 13 pregnancies, one of which resulted in an unspecified malformation (Kaaja et al., 2003; Wide et al., 2004). Meischenguiser et al., (2004) reported on the Argentinian experience of 35 monotherapy exposures with no major congenital anomalies. There was also one cardiac anomaly with 20 cases of OXC polytherapy (Meischenguiser et al., 2004). In the case of the cardiac anomaly, the pregnancy was exposed to 1200 mg OXC and 150 mg phenobarbital. Since the numbers of pregnancies are small, it cannot be known whether the reported physical birth defects were due to OXC or to other factors.

It should be noted that OXC is a derivative of carbamazepine (brand name: Tegretol), a first generation AED and folate antagonist. Carbamazepine is associated with a 6.5% congenital malformation rate and a 1% chance for neural tube defects (Wide et al., 2004). Based on what is known about carbamazepine, maternal serum AFP screening, targeted fetal ultrasound and fetal echocardiogram should be performed in pregnancies treated with OXC.

Topamax (generic name: topiramate)
Topiramate (TPM) is structurally and pharmacologically different from other classes of AED (Ohman et al., 2002). According to the manufacturer, when TPM was administered to pregnant mice, rats and rabbits during organogenesis, increased fetal mortality and teratogenic effects were observed at doses lower than the recommended human dose. Craniofacial and limb malformations were seen most frequently. These malformations were consistent with malformations observed in animals treated with similar medications, the carbonic anhydrase inhibitors. However, since these malformations have not been observed in humans treated with other carbonic anhydrase inhibitors, the manufacturer suggests these effects may be species-specific (Ortho-McNeil Pharmaceutical Communication, 2003).

In the case of TPM, human reproductive data is limited to individual case reports and small case series. A report of 5 pregnant women on TPM polytherapy revealed no congenital malformations at delivery (Öhman et al., 2002). A postmarketing survey by the manufacturer noted that there were 10 pregnancies treated with monotherapy and no congenital malformations identified (Ortho MacNeil Pharmaceutical Communication, 2003).

There is a single case report of an in utero exposure to 1400 mg daily of TPM throughout gestation that resulted in an infant with multiple minor anomalies comprising of hirsutism, third fontanelle, anteverted nares, nail hypoplasia and consistent with the effects seen with first generation AEDs (Hoyme et al., 1998). Additionally, the manufacturer noted that they had received case reports of hypospadias (Ortho MacNeil Pharmaceutical Communication, 2003). No causal relationship could be established with this type of data. Due to the limits in available human reproductive data, fetal risks have not been determined following TPM exposure.

Neurontin (generic name: gabapentin)
A study in mice, rats, and rabbits did not observe any developmental toxicity of gabapentin (Petrere and Anderson, 1994). However, the manufacturer noted that offspring of mice and rats treated with one to four times the recommended human dose of gabapentin (GBP) had delayed ossification of several skeletal bones. Studies on rabbits demonstrated increased fetal loss rates at one-fourth the maximum human dose but no increase in malformations (Pfizer Pharmaceuticals, 2004).

A post-marketing surveillance study of GBP included 11 pregnancy outcomes with first trimester GBP exposure. No congenital abnormalities were observed (Wilton and Shakir, 2002).

The Boston GBP registry collected a combination of retrospective and prospective reports to comment on 51 pregnancies with 44 livebirths. Monotherapy accounted for 33% of the cases, and 81% of the pregnancies were exposed to GBP throughout pregnancy. There were two major congenital anomalies (one complicated with valproate) for a major malformation rate of 4.5% (Montouris, 2003). Due to small sample size, this is considered a very tentative estimate.

Summary
Lamictal is the only second generation AED with sufficient reproductive data suggesting it does not significantly increase the chance for congenital anomalies. Information on possible longterm neurobehavorial effects however is pending.

Antiepileptic Drug Registry
More studies are needed to assess the risks of AED use during pregnancy since discontinuing medication is not an option for most women with epilepsy. The Genetics and Teratology Unit at Massachusetts General Hospital has established the first US hospital-based AED registry. The purpose of the registry is to provide a faster method for establishing the effects of each AED during pregnancy and to provide better counseling and management to pregnant women with epilepsy.

Ascertaining whether a medication has detrimental effects requires large numbers of pregnant women. About 400 women taking a single AED must be enrolled before any study can statistically identify a doubling of the baseline risk for physical birth defects. Many more women are needed to detect more subtle effects of any medication. This AED registry does not release outcome data until statistical significance is reached. Although this has been a source of frustration for health care professionals, the risk of major malformations in various series of 25 birth outcomes has ranged from 3% to 35% for the same AED (Holmes, 2004 OTIS update). Therefore, the registry does not want either to falsely reassure or falsely alarm until the data is statistically significant.

Health care professionals can obtain more information about the AED registry by calling 1-888-233-2334, or visiting http://www.mgh.harvard.edu/aed/ Physicians can refer interested women to these same contacts, but only pregnant women can enroll themselves in the registry.

References
American Academy of Neurology (1998) Practice parameter: management issues for women with epilepsy (summary statement). Report of the Quality Standard Subcommittee of the AAN Neurology 51(4): 944-948.

Bennett GD, et al. (1996) Teratogenicity of carbamazeping-10, 11-epoxide and oxcarbazepine in the SWV mouse. J Pharmacol Exp Ther 279(3): 1237-1242.

Dolk H and McElhatton P (2002) Assessing epidemiological evidence for the teratogenic effects of anticonvulsant medications. J Med Genet 39:243-244.

Friis ML, et al. (1993) Therapeutic experiences with 947 epileptic out-patients in oxcarbazepine treatment. Acta Neurol Scand 87 (3):224-227.

GlaxoSmithKline – Lamotrigine Pregnancy Registry Interim Report. Issued July 2004;1-49.

Holmes LB (2004) Update on the Risk of New Anticonvulsants. OTIS 17th International Conference. Vancouver, BC, Canada.

Holmes LB, et al (2001) The Teratogenicity of Anticonvulsant Drugs. N Engl J Med 344:1132-8.

Kaaja E, et al. (2003) Major malformations in offspring of women with epilepsy. Neurology 60: 575-579.

Marchi NSA, Azoubel R, Tognola WA (2001) Teratogenic effects of lamotrigine on rat fetal brain. Arq Neuropsiquiatr 59(2-B): 362-364.

Meischenguiser R, et al. (2004) Oxcarbazepine in pregnancy: clinical experience in Argentina. Epilepsy and Behavior 5:163-167.

Montouris G (2003) Gabapentin exposure in human pregnancy: results from the Gabapentin Pregnancy Registry. Epilepsy and Behavior 4:310-317.

Morrell M (2003) Reproductive and Metabolic Disorders in Women with Epilepsy. Epilepsia 44 (Suppl. 4)11-20.

Novartis Pharmaceuticals Corporation (2004) Trileptal (oxcarbazepine) package insert.

Öhman I, et al. (2002) Topiramate kinetics during delivery, lactation, and in the neonate: preliminary observations. Epilepsia 43 (10): 1157-1160.

Petrere JA and Anderson JA (1994) Developmental toxicity studies in mice, rats, and rabbits with the anticonvulsant gabapentin. Fundam Appl Toxicol 23:585-9.

Contributors
Amber Pakilit, BS
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.

PDF Version

Common Summertime Exposures During Pregnancy

Volume 11, No.2, June 2004

Sunscreen
Heightened attention to health risks from sun overexposure has increased the use of sunscreens, both in the general population as well as in pregnancy (Benson, 2000). Sunscreens can come in various formulations (creams, lotions, sprays, etc.) and are also present in many cosmetic products. A few common sunscreen ingredients include oxybenzone, avobenzone (Parsol 1789), p-Aminobenzic acid (PABA), octyl methoxycinnimate, and octocrylene (Benson, 2000).

Although there is a lack of specific human pregnancy teratology studies, reassurance can be gathered from their common use during pregnancy and the absence of adverse pregnancy reports.

Additionally, some high dose experimental animal studies are available. For example, Stroeva, et al. (1998) noted that when pregnant rats were directly injected with PABA, there was no increase in congenital malformations. Octocrylene has also been studied in pregnant rats and rabbits. When orally administered to rats and topically applied to rabbits, no teratogenic effects were seen (Odio et al, 1994).

The route of administration is also an important principle in teratology. Since sunscreen is topically applied to the skin, systemic absorption (and therefore fetal exposure) is expected to be low. Depending on the particular ingredient, it is estimated that only 1-10% of topically applied sunscreen is absorbed into the blood system (Benson, 2000).

There is a medical benefit of sunscreen use in protecting individuals from the damaging effects of the sun. For this reason, sunscreen has been used for many years as a method of preventative health care (Benson, 2000). It is important that a pregnant woman, like all women, protect herself from overexposure to the sun. Although specific pregnancy studies are limited, conservative, but appropriate use of sunscreens on exposed areas of skin should not be a concern for the developing fetus.

Self-tanners
Self-tanning creams, lotions, and sprays typically contain an active ingredient called dihydroxyacetone (DHA) at concentrations of 3-5% (Levy, 1992). When applied topically, DHA temporarily enhances pigmentation of the skin. Due to the topical route of administration, systemic absorption is expected to be low. One in vitro study estimated that while 22% of the applied dose penetrated human skin, only 0.5% of this application would be systemically absorbed (Yourick et al, 2004).

As with sunscreen, while there are no adverse pregnancy reports, there are also no human pregnancy studies on DHA. However, there are no experimental animal studies on DHA either. Since self-tanners do not have the medical benefit that sunscreen has, a woman may feel more comfortable erring on the side of caution and deciding to avoid the use of self-tanners during pregnancy. If a woman does choose to use tanning creams during pregnancy, it is important to remember that they do not provide protection from the sun and that the small amount of expected absorption can be further decreased with less frequent use and with application to smaller areas of the body.

West Nile Virus
In 1999, New York City was the cite of an outbreak of encephalitis and meningitis caused by the West Nile Virus (WNV), originally described in 1937 in Uganda (Alpert et al, 2003; Chappa et al, 2003). The disease is transmitted to humans primary through the bites of infected mosquitoes.

Symptoms of infection typically develop within 3-14 days after being bitten, and can vary from mild to severe. Many people infected with WNV do not develop any symptoms. The mild form of WNV typically presents as a sudden fever, often with nausea, vomiting, eye pain, skin rash, headache, or myalgia. Less than 1% of cases result in more serious neurological disease, with advanced age being a risk factor. Clinical information can be located online at http://www.cdc.gov/ncidod/dvbid/westnile/resources/fact_sheet_clinician.htm

In 2002, the Centers for Disease Control and Prevention (CDC) reported the first case of intrauterine transmission from a woman who had WNV encephalitis during the 27th week of pregnancy (Alpert et al, 2003). The child was born with both ocular and neurologic complications. Serum studies on the infant indicated the presence of WNV-specific antibodies. Although this case demonstrated intrauterine WNV infection, no causal relationship between WNV and the congenital anomalies was established.

However, after this case was reported, a registry was formed by the CDC and local state and health departments to follow birth outcomes among women with WNV illness during pregnancy (O’Leary, 2004). During 2002, three additional infants of mothers infected with WNV were born full term with normal appearance and negative laboratory studies. Since 2003, the registry has identified 74 women who acquired WNV illness while pregnant. Preliminary data for 49 known outcomes includes 42 livebirths, 5 first trimester miscarriages, and two elective terminations. Of the 42 livebirths, specimens were available from 29 to check for WNV antibodies. There was evidence of intrauterine infection in 1 infant who did not have clinical evidence of illness.

Additionally, in three other cases with suspected, but unconfirmed intrauterine infection, there was 1 infant fatality secondary to lissencephaly and superimposed WNV infection, one case of a neonatal rash that resolved, and one case of a neonatal rash with fever (O’Leary, 2004).

If WNV is diagnosed during pregnancy, ultrasound at 2-4 weeks post symptoms can evaluate the fetus for signs of viral infection (CDC, 2004). While amniotic fluid could be tested for evidence of WNV infection, the sensitivity, specificity, and predictive value are not known, nor is the clinical consequence of fetal infection (CDC ,2004). Physicians aware of instances of WNV in pregnancy are asked to contact their state health department or the CDC. A list of follow-up clinical newborn evaluations can be found online at http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5307a4.htm

In an attempt to minimize the risk of contracting WNV during pregnancy, the CDC recommends that women who live in areas with WNV-infected mosquitoes use insect repellant over their clothes and on exposed skin, and when possible avoid outdoors during dawn and dusk, the peak hours of mosquito activity (CDC, 2004).

DEET
DEET is an acronym for Diethyltoluamide, which has been marketed in the United States since the late 1950s. According to the CDC, it is the most effective repellant for the prevention of mosquito bites. DEET is a common component of many insect repellents such as “Off”. In a study investigating the absorption of DEET following dermal application in a group of male volunteers, it was found that on average, only 5.6% of DEET was systemically absorbed at undiluted concentrations (Selim et al, 1995).

DEET has been reported to produce dermatitis in sensitive adult individuals (Koren et al, 2003). Initial concern about DEET in pregnancy stemmed from case reports of seizures following misuse of DEET in children (Zadikoff et al, 1979). However, causation related to DEET was not proven. Additionally, a more recent study did not find a higher incidence of severe adverse events in children compared to adults (Koren et al, 2003).

Very high dose animal studies in rats and rabbits (undiluted DEET administered via a stomach tube) found no increased incidence of congenital anomalies (Schoenig GP et al, 1994). At the highest doses, decreased fetal body weights were found. However, the latter may be related to the toxicity of the mother animal and reduced maternal food consumption.

There is an isolated case report of an adverse pregnancy outcome in association with the daily use of DEET throughout the pregnancy, but cause and effect were not established because there were additional medicine exposures (Schaefer et al, 1992).

Unfortunately, there is no specific controlled data available on human first trimester exposure to DEET to confirm the reassuring animal data. However, there is a controlled human pregnancy study that evaluated DEET exposure during the second and third trimesters of pregnancy.

A double-blinded, randomized, therapeutic trial of insect repellents for the prevention of malaria was conducted in Thailand (McGready et al, 2001). 897 women between three and seven months of pregnancy participated and were randomly allocated to either receive a 20% DEET solution and thanaka (a common local cosmetic paste used as a carrier for the repellent), or thanaka alone. 449 women were exposed to the DEET and thanaka combination, while 448 women were exposed to the thanaka alone. 741 live born singletons were available for follow-up studies. In 50 women randomly selected who received DEET, 4 (8%) showed evidence of DEET in the cord blood, indicating that placental transfer can occur. However, there were no differences in survival or growth parameters between the two groups, and for those infants followed for the first year of life (81%), no differences were noted in neurological development.

It should be noted that repellents containing a higher concentration of DEET provide longer-lasting protection but not more effective protection. A repellent therefore should be chosen with the lowest concentration needed for the amount of time spent outdoors. The CDC notes that a product containing 20% DEET provides almost 4 hours of protection while a product with 6.65% DEET provides almost 2 hours of protection and a products with 4.75% DEET provides roughly 1 and one half hour of protection.

Patient information on the use of DEET during pregnancy can be found at www.otispregnancy.org under the fact sheet listings.

In summary, both experimental animal studies and one controlled human study during the second and third trimesters regarding the use of DEET found no increase in congenital anomalies.

Limiting the number and amount of DEET applications should reduce systemic absorption and therefore reduce fetal exposure. Other recommendations listed earlier include wearing protective clothing (i.e. long sleeves, socks) when possible and then spraying DEET over clothing, rather than directly on skin, not spraying directly onto abraided skin, and avoiding outside activities during peak hours of mosquito exposure. After returning indoors, the CDC recommends washing any treated skin with soap and water. They also recommend removing any items outside and around the home that contain standing water where mosquitoes can lay their eggs.

References
Abdel-Rahman A, et al. (2004) Neurological deficits induced by malathion, DEET, and permethrin, alone or in combination in adult rats. J Toxicol Environ Health A.67(4):331-56.

Alpert SG, et al. (2003) Intrauterine West Nile Virus: ocular and systemic findings. Am J Ophthalmol 136 (4):733-735.

Benson H (2000) Assessment and clinical implications of absorption of sunscreens across skin. Am J Clin Dermatol 1(4):217-224.

Chapa JB, et al. (2003) Wile Nile encephalitis during pregnancy. Obstet and Gynecol 102 (2):229-231.

CDC (2004) Interim guidelines for the evaluation of infants born to mothers infected with WNV during pregnancy. MMWR Weekly 53 (7); 154-157.

CDC WNV Chapter http://www.cdc.gov/ncidod/dvbid/westnile/index.htm

Hayden C, et al. (1997) Systemic absorption of sunscreen after topical application. Lancet 350 (9081):863-4.

Koren G, et al. (2003) DEET-based insect repellents: safety implications for children and pregnant and lactating women. CMAJ 169 (3):209-211.

Levy SB (1992) Dihydroxyacetone-containing sunless or self-tanning lotions. J Am Acad Dermato 27: 989-993.

Odio MR, et al. (1994) Evaluation of subchronic (13 week), reproductive, and in vitro genetic toxicity potential of 2-ethylhexyl-2-

cyano-3,3-diphenyl acrylate (octocrylene). Fundam Appl Toxicol 22:355-368.

O’Leary D (2004) Fifth National Conference on WNV in US: Denver, Colorado.

Schaefer C, et al. (1992) Intrauterine diethyltoluamide exposure and fetal outcome. Reprod Toxicol 6 (2):175-6.

Schoenig GP, et al (1994) Teratologic evaluations of N, N-diethyl-M-toluamide (DEET) in rats and rabbits. Fundam Appl Toxicol 23 (1):63-9.

Selim S, et al. (1995) Absorption, metabolism, and excretion of N,N-diethyl-M-touamide following dermal application to human volunteers. Fundam Appl Toxicol 25:95-100.

Stroeva OG (1998) Effect of para-aminobenzoic acid on the development of rat embryos when applied to pregnant females. Ontogenez 29(6):444-9.

Zadikoff CM (1979) Toxic encephalopathy associated with use of insect repellant. J Pediatr 95 (1):140-142.

Contributors
Sarah Lewis, BS
Genetic Counseling Intern
Mara Gaudette, MS, CGC
Coordinator, Illinois Teratogen Information Service
Eugene Pergament, MD, PhD, FACMG
Northwestern Reproductive Genetics, Inc.